Processing math: 45%
Research article Special Issues

Optimization of surgical protocol for laser iridotomy based on the numerical simulation of aqueous flow

  • Primary angle-closure glaucoma (PACG) is a major cause of blindness worldwide, with a particularly high prevalence in Asian populations. Laser iridotomy (LI) has been the standard therapeutic modality for treating PACG to avoid blindness. However, the complex structure of the eyeball, the aqueous fluidity, and the limitation of detecting equipment will cause difficulty in surgery and the probability of complications. Numerical simulation was conducted to investigate aqueous humor (AH) flow under different physiological structures before and after laser surgery. When the anterior chamber depth decreases from 2.8 mm to 2.0 mm (caused by angle-closure glaucoma), the maximum velocity of natural convection is doubled, and the pressure difference between the posterior and anterior chambers increases by 20%. Therefore, a shallow anterior chamber depth is crucial for the accurate investigation of glaucoma. Pupil block sharply increases the intraocular pressure (IOP). When the gap between the lens and iris decreases from 10 μm to 0.5 μm, P between the posterior and anterior chambers is approximately 37 times higher than before, thereby damaging intraocular tissues. LI can effectively reduce the IOP caused by pupil block, but the velocity of AH after operation is 40 times the normal condition, and the increased corneal shear stress could lead to corneal damage, which can be solved by adjusting the incident angle of laser beam On the basis of the allowable angle range of surgical equipment and the effect of different incident angles on the cornea and iris, the optimum angle of laser drilling is 45°.

    Citation: Yibo Zhao, Bin Chen, Dong Li. Optimization of surgical protocol for laser iridotomy based on the numerical simulation of aqueous flow[J]. Mathematical Biosciences and Engineering, 2019, 16(6): 7405-7420. doi: 10.3934/mbe.2019370

    Related Papers:

    [1] Neda Askari, Hassan Momtaz, Elahe Tajbakhsh . Acinetobacter baumannii in sheep, goat, and camel raw meat: virulence and antibiotic resistance pattern. AIMS Microbiology, 2019, 5(3): 272-284. doi: 10.3934/microbiol.2019.3.272
    [2] Aynur Aybey, Elif Demirkan . Inhibition of Pseudomonas aeruginosa biofilm formation and motilities by human serum paraoxonase (hPON1). AIMS Microbiology, 2016, 2(4): 388-401. doi: 10.3934/microbiol.2016.4.388
    [3] Liang Wang, Zuobin Zhu, Huimin Qian, Ying Li, Ying Chen, Ping Ma, Bing Gu . Comparative genome analysis of 15 clinical Shigella flexneri strains regarding virulence and antibiotic resistance. AIMS Microbiology, 2019, 5(3): 205-222. doi: 10.3934/microbiol.2019.3.205
    [4] Maureen U. Okwu, Mitsan Olley, Augustine O. Akpoka, Osazee E. Izevbuwa . Methicillin-resistant Staphylococcus aureus (MRSA) and anti-MRSA activities of extracts of some medicinal plants: A brief review. AIMS Microbiology, 2019, 5(2): 117-137. doi: 10.3934/microbiol.2019.2.117
    [5] Kamila Tomoko Yuyama, Thaís Souto Paula da Costa Neves, Marina Torquato Memória, Iago Toledo Tartuci, Wolf-Rainer Abraham . Aurantiogliocladin inhibits biofilm formation at subtoxic concentrations. AIMS Microbiology, 2017, 3(1): 50-60. doi: 10.3934/microbiol.2017.1.50
    [6] Joseph O. Falkinham . Mycobacterium avium complex: Adherence as a way of life. AIMS Microbiology, 2018, 4(3): 428-438. doi: 10.3934/microbiol.2018.3.428
    [7] Itziar Chapartegui-González, María Lázaro-Díez, Santiago Redondo-Salvo, Elena Amaro-Prellezo, Estefanía Esteban-Rodríguez, José Ramos-Vivas . Biofilm formation in Hafnia alvei HUMV-5920, a human isolate. AIMS Microbiology, 2016, 2(4): 412-421. doi: 10.3934/microbiol.2016.4.412
    [8] Tatyana V. Polyudova, Daria V. Eroshenko, Vladimir P. Korobov . Plasma, serum, albumin, and divalent metal ions inhibit the adhesion and the biofilm formation of Cutibacterium (Propionibacterium) acnes. AIMS Microbiology, 2018, 4(1): 165-172. doi: 10.3934/microbiol.2018.1.165
    [9] Yunfeng Xu, Attila Nagy, Gary R. Bauchan, Xiaodong Xia, Xiangwu Nou . Enhanced biofilm formation in dual-species culture of Listeria monocytogenes and Ralstonia insidiosa. AIMS Microbiology, 2017, 3(4): 774-783. doi: 10.3934/microbiol.2017.4.774
    [10] Stephen H. Kasper, Ryan Hart, Magnus Bergkvist, Rabi A. Musah, Nathaniel C. Cady . Zein nanocapsules as a tool for surface passivation, drug delivery and biofilm prevention. AIMS Microbiology, 2016, 2(4): 422-433. doi: 10.3934/microbiol.2016.4.422
  • Primary angle-closure glaucoma (PACG) is a major cause of blindness worldwide, with a particularly high prevalence in Asian populations. Laser iridotomy (LI) has been the standard therapeutic modality for treating PACG to avoid blindness. However, the complex structure of the eyeball, the aqueous fluidity, and the limitation of detecting equipment will cause difficulty in surgery and the probability of complications. Numerical simulation was conducted to investigate aqueous humor (AH) flow under different physiological structures before and after laser surgery. When the anterior chamber depth decreases from 2.8 mm to 2.0 mm (caused by angle-closure glaucoma), the maximum velocity of natural convection is doubled, and the pressure difference between the posterior and anterior chambers increases by 20%. Therefore, a shallow anterior chamber depth is crucial for the accurate investigation of glaucoma. Pupil block sharply increases the intraocular pressure (IOP). When the gap between the lens and iris decreases from 10 μm to 0.5 μm, P between the posterior and anterior chambers is approximately 37 times higher than before, thereby damaging intraocular tissues. LI can effectively reduce the IOP caused by pupil block, but the velocity of AH after operation is 40 times the normal condition, and the increased corneal shear stress could lead to corneal damage, which can be solved by adjusting the incident angle of laser beam On the basis of the allowable angle range of surgical equipment and the effect of different incident angles on the cornea and iris, the optimum angle of laser drilling is 45°.


    The present work is motivated by the study of nonlocal peridynamics models initially proposed by Silling in [31]. In particular, the state-based peridynamics model given in [31,32,33] postulates that the total strain energy for constitutively linear, isotropic solid undergoing deformation is given by

    Wρ(u)=βΩ(Dρ(u)(x))2dx+αΩΩρ(xx)(D(u)(x,x)1dDρ(u)(x))2dxdx (1.1)

    where ΩRd is a bounded domain occupied by the solid material, the kernel ρ(|ξ|) is a nonnegative locally integrable and radial weight function that measures the interaction strength between material particles at position x and x, u is a displacement field, D(u) is a rescaled nonlocal operator on u defined by [11]

    D(u)(x,x)=(u(x)u(x))|xx|(xx)|xx|=(u(x)u(x))T(xx)|xx|2, (1.2)

    representing a (unit-less) linearized nonlocal strain [32] and the operator Dρ is a nonlocal linear operator (a weighted version of D [11,12]), called 'nonlocal divergence', which is defined as

    Dρ(u)(x):=p.v.Ωρ(xx)D(u)(x,x)dx (1.3)

    which is a means of incorporating the effect of the collective deformation of a neighborhood of x into the model. The positive constants α and β are proportional to the shear and bulk moduli of the material, respectively. The quadratic energy in (1.1) is a generalization of the bond-based model that was introduced in [31] and studied in [1,13,15,23,37] that takes in to account the linearized strain due to the dilatation and the deviatoric portions of the deformation. Mathematical analysis of linearized peridynamic models have been extensively studied in [10,11,12,13,15,23,24,37] along with results geared towards nonlinear models in [3,4,9,14,19,25].

    For ρL1loc, it is not difficult to show (see Lemma 2.3 below) that the energy space associated with the energy functional Wρ, {uL2(Ω;Rd):Wρ(u)<}, is precisely

    Sρ,2(Ω)={uL2(Ω;Rd):|u|2Sρ,2<}, (1.4)

    where the seminorm |u|Sρ,2 is defined by

    |u|2Sρ,2:=ΩΩρ(yx)|(u(y)u(x))|yx|(yx)|yx||2dydx.

    Notice that |u|Sρ,2=0, if and only if u is an infinitesimal rigid vector field. We denote the class of infinitesimal rigid displacements by

    R:={u:u(x)=Bx+v,BT=B,vRd}.

    It has been shown in [23,25] that Sρ,2(Ω) with the natural norm

    uSρ,2=(u2L2+|u|2Sρ,2)1/2

    is a separable Hilbert space. In the event that ρ(ξ)|ξ|2L1loc(Rd), then the space Sρ,2(Ω) coincides with L2(Ω,Rd). Otherwise, Sρ,2(Ω) is a proper subset of L2(Ω,Rd) that is, nevertheless, big enough to contain W1,2(Ω;Rd) and there exists a constant C=C(d,2,Ω)>0 such that

    |u|2Sρ,2CSym(u)2L2ρL1(R),uW1,2(Ω;Rd)

    where Sym(u)=12(u+uT) is the symmetric part of the gradient u.

    Under the additional assumption that ρ is positive in a small neighborhood of the origin, it is shown in [23, Theorem 1], via an application of Lax-Milgram, that for any applied load fL2(Ω;Rd), the potential energy

    Eρ(u)=Wρ(u)Ωfudx (1.5)

    has a minimizer over any weakly closed subset V of Sρ,2(Ω) such that VR={0}. See also [25, Theorem 1.1] for the more general convex energies of p-growth.

    The analysis of the convergence of variational problems of the type in (1.5) associated with a sequence of parameterized kernels has garnered a lot of attention in recent years. Namely, if we have a sequence of locally integrable radial kernels ρn, how do the associated potential energies Eρn, as well as their minimizers behave as n? Clearly, this will depend first on the behavior of the convergence properties of the sequence of kernels. In fact, it is shown in [25] that if the sequence of L1 kernels {ρn} converge in the sense of measures to a measure with atomic mass at 0 (Dirac-measure at 0) and for each n, r2ρn(r) is nonincreasing, then the sequence {Eρn} variationally converges to the classical Navier-Lamé potential energy E0 given by

    E0(u)=μΩ|Sym(u)|2dx+λ2Ω(div(u))2dxΩfudx,

    where μ and λ are constants that can be expressed in terms of α and β. This is what is called nonlocal-to-local convergence and the result is used as a rigorous justification that state-based peridynamics modeling recovers the classical linearized elasticity models in the event of vanishing nonlocality.

    In this paper, we consider another type of convergence of sequence of kernels and study the behavior of the associated energy functionals, which leads to nonlocal-to-nonlocal convergence. More specifically, suppose we are given a nonnegative kernel ρL1loc(Rd) with the property that

    ρ is radial ,ρ(ξ)>0 for ξ is close to 0,|ξ|2ρ(|ξ|) is nonincreasing in |ξ|, (1.6)

    and

    limδ0δ2(Bδρ(ξ)dξ)1=0. (1.7)

    and consider a sequence of nonnegative, radial kernels {ρn}nN each satisfying (1.6) and that

    ρnρa.e. and ρnρa.e. in Rd. (1.8)

    It then follows that ρnρ strongly in L1loc(Rd) as n. We will establish a clear connection between the sequence of energies {Eρn} and Eρ. Most importantly, we will show that minimizers of the energies Eρn over an admissible class will converge to a minimizer of Eρ over the same admissible class. The notion of variational convergence we use is Γ-convergence (see [21]) which we define below. The advantage of Γ-convergence is that under the additional assumption of equicoercivity of the functionals it implies the convergence of minimizers as well [21, Theorem 7.8 and Corollary 7.202].

    Definition 1.1. Suppose that ¯En:L2(Ω;Rd)R{}, 1n. We say that the sequence ¯En Γ- converges to ¯E in the L2-topology if and only if

    a) for every sequence {un}L2(Ω;Rd) with unu in L2(Ω;Rd), as n, we have ¯E(u)lim infn¯En(un),

    b) and for every uL2(Ω;Rd) there exists a recovery sequence unu in L2(Ω;Rd), such that ¯E(u)=limn¯En(un).

    The following is one of the main results of the paper on the variational limit of the nonlocal functionals {¯En}.

    Theorem 1.2. Suppose ρ and {ρn} satisfies (1.6)–(1.8). The sequence of functionals ¯En Γconverges in the strong L2(Ω;Rd) topology to the functional ¯Eρ, where the extended functionals {¯En(u)}n are defined as

    ¯En(u)={Eρn(u),if uSρn,2(Ω),,if uL2(Ω;Rd)Sρn,2(Ω), (1.9)

    where Sρn,2(Ω) and {Eρn}n< are defined as before in (1.4) and (1.5), respectively, where ρ is replaced by ρn. The extended functional ¯Eρ is similarly defined. Moreover, if V is a weakly closed subset of L2(Ω;Rd) such that VR={0}, and for each n, un minimizes Eρn over VSρn,2(Ω), then the sequence {un} is precompact in L2(Ω;Rd) with any limit point belonging to Sρ,2(Ω) and minimizing Eρ over VSρ,2(Ω).

    Although the discussion above is focused on the case of quadratic peridynamic energies for ease of explaining the main ideas, the result can naturally be extended to small strain nonlocal nonlinear peridynamic models with p-growth, for p2, that have been introduced in [33] and whose variational analysis investigated in [25].

    We will prove Theorem 1.2 in the sections that follow. But we would like to highlight that this result has an important implication in the numerical approximation of minimizers of Eρ over an admissible class. Indeed, compactness results have been quite useful for analyzing numerical approximations of nonlocal problems in various contexts such as [8,34,35]. In the context discussed in this work, let us take for an example that ρ(ξ)=1|ξ|d+2(s1) for s(0,1). This kernel satisfies (1.6) and (1.7). It is also clear that ρ(ξ)|ξ|2 is not integrable on any bounded domain containing 0. In the event Ω has a smooth boundary, the energy space Sρ,2(Ω) coincides with the fractional Sobolev space Hs(Ω;Rd) see [27,30]. In particular, if 1/2<s<1, then all functions in Sρ,2(Ω) have continuous representative. Now, if VSρ,2(Ω) is a weakly closed subset of L2(Ω;Rd) such that VR={0}, a minimizer of Eρ over V exists (and will be in Hs(Ω;Rd)). The analysis of the existence and uniqueness of the minimizer u of this quadratic energy can also, equivalently, be found by solving the corresponding Euler-Lagrange equation. The latter gives us a way of numerically solving the solution by writing it first in the weak form and then applying the Galerkin approach of choosing a finite-dimensional subspace MV to solve for a projected solution of u on M. Notice that for s(0,1/2) the finite-dimensional subspace M can contain discontinuous functions, while for s(1/2,1), all the elements of M must be continuous in order for M to be conforming, that is, for MSρ,2(Ω). In the latter case, if one wants to employ the advantageous discontinuous Galerkin approximation, which is now nonconforming, one needs to find an effective way to implement it to the model problem. The result in Theorem 1.2 will allow us to develop approximation schemes by solving a sequence of Euler-Lagrange equations of modified energies. To demonstrate this, define the sequence of kernels

    ρn(ξ)={ρ(ξ)if ρ(ξ)n|ξ|2n|ξ|2if ρ(ξ)n|ξ|2.

    It is not difficult to check that, for each n, ρn satisfies (1.6), (1.8), and that the functions ρn(ξ)|ξ|2L1loc(Rn) are just truncations of the fractional kernel |ξ|d2s (at level n). As discussed before, the energy space associated with Eρn will coincide with L2(Ω;Rd) and a unique minimizer unVSn of Eρn exists. Since the admissible space is a subspace of L2(Ω;Rd) that avoids nontrivial infinitesimal rigid displacements, we may use discontinuous finite element spaces, denoted by Mn,h, for the standard conforming Galerkin approximation to the solution of the Euler-Lagrange equation associated with the energy Eρn. This, in turn, can be viewed as a nonconforming discontinuous Galerkin scheme to numerically solve the original Euler-Lagrange equations when the discretization parameter h goes to zero and at the same time the truncation level n goes to infinity. Intuitively, for large n, by Theorem 1.2, u is approximated by un (in the L2 norm), and then un will be numerically approximated by un,hMn,h. The proper convergence analysis of this nonconforming numerical scheme as h0 and n simultaneously has been carried out in [34] in the special case of scalar nonlocal problems when the subspace V is the set of scalar-valued functions uL2(Ω;Rd) such that u vanishes outside of a fixed set Ω which is compactly contained in Ω. The analysis in [34] makes use of the framework of asymptotically compatible schemes for parameter-dependent problems first developed in [35] and the vanishing of the functions in the admissible class around the volumetric-boundary ΩΩ is crucial for employing certain compactness arguments. To extend the convergence analysis in [34] to the case of a system of strongly coupled nonlocal equations, the variations problems associated with (1.5), solved over any admissible set that does not include infinitesimal rigid vector fields, Theorem 1.2 as well as some of the compactness results we prove in this paper will be crucial. Such analysis on nonconforming discontinuous Galerkin numerical schemes to systems of nonlocal equations under discussion will be carried out in a future work.

    Although it is beyond the scope of this work, in passing, we would like to note that this way of developing a nonconforming numerical scheme is also applicable to fractional PDEs where singular kernels are more common [5,36]. The idea is the same where we use less singular kernels with truncation both at origin and at infinity to do approximation of fractional PDEs. In this case, sequential compactness of nonlocal spaces associated with the truncated fractional kernel together with the compact embedding of fractional Sobolev spaces in Lp can be used to carry out the analysis of the resulting asymptotically compatible schemes [35,36].

    The proof of Theorem 1.2 fundamentally depends on some structural properties of the nonlocal space Sρ,2(Ω), chief among them are compact embedding into L2(Ω;Rd) and a Poincaré-type inequality, which we will establish in this paper. In fact, these properties remain true even for the spaces Sρ,p(Ω), where for 1p<,

    Sρ,2(Ω)={uLp(Ω;Rd):|u|Sρ,p<},

    and

    |u|pSρ,p=ΩΩρ(yx)|(u(y)u(x))|yx|(yx)|yx||pdydx

    gives a semi norm. It is shown in [20,25] that, for any 1p<, Sρ,p(Ω) is a separable Banach space with the norm

    uSρ,p=(upLp+|u|pSρ,p)1/p,

    and is reflexive if 1<p< and a Hilbert space for p=2. If |ξ|pρ(ξ)L1loc(Rd), then a simple calculation shows that Sρ,p(Ω)=Lp(Ω;Rd). On the other hand, in the case where |ξ|pρ(ξ)L1loc(Rd), Sρ,p(Ω) is a proper subset of Lp(Ω;Rd). Under some extra assumptions on the kernel ρ, the space is known to support a Poincaré-Korn type inequality over subsets that have trivial intersections with R. These functional analytic properties of the nonlocal space can be used to demonstrate the well-posedness of some nonlocal variational problems using the direct method of calculus of variations, see [25] for more discussions.

    As in the case of p=2, we assume that for a given 1p<,

    ρ is radial,ρ(r)>0 for r is close to 0, and rpρ(r) is nonincreasing in r, (1.10)

    and

    limδ0δp(Bδ(0)ρ(ξ)dξ)1=0. (1.11)

    We now state the compactness result whose proof is one of the main objectives of the present work.

    Theorem 1.3 (Lp compactness). Let 1p< and let ρL1loc(Rd) be nonnegative and satisfying (1.10) and (1.11). Suppose also that ΩRd is a domain with Lipschitz boundary. Then Sρ,p(Ω) is compactly embedded in Lp(Ω;Rd). That is, any bounded sequence {un} in Sρ,p(Ω) is precompact in Lp(Ω;Rd). Moreover, any limit point is in Sρ,p(Ω).

    The condition given by (1.11) requires ρ to have an adequate singularity near 0. A straightforward calculation shows that the kernels satisfying (1.11) include ρ(ξ)=|ξ|(d+p(s1)), for any p[1,), and any s(0,1), and ρ(ξ)=|ξ|pdln(|ξ|). It is no surprise that (1.11) is violated if |ξ|pρ(ξ) is a locally integrable function (and therefore, Sρ,p(Ω)=Lp(Ω;Rd)), and in fact, in this case

    lim infδ0δp(Bδ(0)ρ(ξ)dξ)1=,

    see [25]. It is not clear whether condition (1.11) is necessary for compact embedding even for the class of kernels that are radial and nonincreasing. There are radial kernels with the property that |ξ|pρ(ξ) is (locally) nonintegrable, and

    limδ0δp(Bδ(0)ρ(ξ)dξ)1=c0>0

    for which we do not know whether there is a compact embedding Sρ,p(Ω) into Lp(Ω;Rd). One such kernel is ρ(ξ)=|ξ|pd. Nevertheless, we can prove that the associated space Sρ,p(Ω) is compact in the Lploc topology, a result which we will state and prove in the appendix.

    For scalar fields, compactness results like those stated above are commonplace for spaces corresponding to special kernels such as the standard fractional Sobolev spaces. In [22, Lemma 2.2], for more general radial and monotone decreasing kernels ρ, condition (1.11) is shown to be sufficient for the compact embedding of the space

    {fL2(Ω):ΩΩρ(yx)|f(y)f(x)|2|yx|2<}

    in L2(Ω). The statement is certainly true for any 1p<. The proof of [22, Lemma 2.2] actually relies on and modifies the argument used to prove another type of compactness result by Bourgain, Brezis and Mironescu in [6, Theorem 4] that applies criteria involving a sequence of kernels. The argument of [6] uses extensions of functions to Rd where the monotonicity of ρ is used in an essential way to control the semi-norm of the extended functions by the original semi-norm. That is, let us introduce a sequence of radial functions ρn satisfying

    n1,ρn0,Rdρn(ξ)dξ=1,and limn|ξ|>rρn(ξ)dξ=0,r>0. (1.12)

    Assuming that for each n, ρn is nonincreasing, and if

    supn1ΩΩρn(yx)|fn(y)fn(x)|p|yx|p<, (1.13)

    then {fn} is precompact in Lp(Ω), which is the result of [6, Theorem 4] obtained by showing that (1.13) makes it possible to apply a variant of the Riesz-Fréchet-Kolomogorov theorem [7]. In [22, Lemma 2.2], for a fixed ρ, the condition (1.11) is used to replace the role played by the condition (1.12). In [28, Theorem 1.2], the same result as in [6, Theorem 4] was proved by dropping the monotonicity assumption on ρn for d2. In addition, the proof in [28] avoids the extension of functions to Rd but rather shows that the bulk of the mass of each fn, that is Ω|fn|p, comes from the interior and quantifies the contribution near the boundary. As a consequence, if (1.13) holds, then as n there is no mass concentration or leak at the boundary, two main causes of failure of compactness. The compactness results were applied to establish some variational convergence results in [29]. Clearly if one merely replaces scalar functions in (1.13) by vector fields, both compactness results [6, Theorem 4] and [28, Theorem 1.2] will remain true. It turns out the results will remain valid for vector fields even under a weaker assumption. Indeed, following the argument [6, Theorem 4] and under the monotonicity assumption that for n, ρn is nonincreasing, it was proved in [20, Theorem 5.1] that if un is a bounded sequence of vector fields satisfying

    supn1ΩΩρn(yx)|D(un)(x,y)|pdydx< (1.14)

    then {un} precompact in the Lploc(Ω;Rd) topology with any limit point being in W1,p(Ω;Rd) when 1<p<, and in BD(Ω) when p=1. Here, BD(Ω) is the space of functions with bounded deformation. Later, again under the monotonicity assumption on ρn, but using the argument of [28, Theorem 1.2] instead, it was proved in [25, Proposition 4.2] that in fact, (1.14) implies that {un} is precompact in the Lp(Ω;Rd) topology. In this paper, we will prove a similar result relaxing the requirement that ρn is a Dirac-Delta sequence.

    Theorem 1.4. Let ρL1loc satisfy (1.10) and (1.11). For each n, ρn is radial and ρn satisfies (1.10) and that

    ρn0,ρnρ,weakly in L1loc(Rd),and ρncρ

    for some c>0. Then, if {un} is a bounded sequence in Lp(Ω;Rd) such that (1.14) holds, then {un} is precompact in Lp(Ω;Rd). Moreover, any limit point is in Sρ,p(Ω).

    A natural by-product of Theorem 1.4 is the Poincaré-Korn type inequality stated below.

    Corollary 1.5 (Poincaré-Korn type inequality). Suppose that 1p< and V is a weakly closed subset of Lp(Ω;Rd) such that VR={0}. Let ρL1loc satisfies (1.10) and (1.11). Let ρn be a sequence of radial functions, and for each n, ρn satisfies (1.10) and that

    ρn0,ρnρ,weakly in L1loc(Rd), and ρncρ

    for some c>0. Then there exist constants C>0 and N1 such that

    Ω|u|pdxCΩΩρn(yx)|(u(y)u(x))|yx|(yx)|yx||pdydx (1.15)

    for all uVLp(Ω;Rd) and nN. The constant C depends only on V,d,p, ρ, and the Lipschitz character of Ω.

    We note that the Poincaré-Korn-type inequality has been proved for Dirac-Delta sequence of kernels ρn [23,25]. The corollary extends the result to sequence of kernels that weakly converge to a given function ρ satisfying (1.10) and (1.11).

    The rest of the paper is devoted to prove the main results and it is organized as follows. We prove Theorem 1.2 in Section 2. Theorem 9.1 and Proposition 3.5 are proved in Section 3. The proof of Theorems 1.3 and 1.4 and Corollary 1.5 are presented in Section 4. Further discussions are given at the end of the paper.

    In this section we will prove the Γ-convergence of the sequence of energies Eρn defined in (1.5). The proof relies on a sequence of results on the limiting behavior of functions as well as the action of operators. To that end, we assume that ρ and {ρn} satisfy (1.6)–(1.8) throughout this section. We begin with the convergence properties of the nonlocal divergence operator.

    Lemma 2.1. Suppose that unu strongly in L2(Ω;Rd), uSρ,2(Ω), and that {Dρn(un)} is uniformly bounded in L2(Ω). Then Dρn(un)Dρ(u) weakly in L2(Ω).

    Proof. We recall the nonlocal integration by parts formula ([25,26]) that for any vW1,20(Ω)

    ΩDρn(un)v(x)dx=ΩGn(v)(x)un(x)dx

    where Gn(v)(x) is the nonlocal gradient operator

    Gρn(v)(x)=p.v.Ωρn(yx)v(y)+v(x)|yx|yx|yx|dy.

    Now for vC1c(Ω), we may rewrite the nonlocal gradient as

    Gρn(v)(x)=Ωρn(yx)v(y)v(x)|yx|yx|yx|dy+2Ωρn(yx)v(x)|yx|yx|yx|dy.

    and estimate as [26, Corollary 2.4],

    Gρn(v)L3ρnL1vLCρL1vL.

    Also, it is not difficult to show that for all xΩ, Gρn(v)(x)Gρ(v)(x). This follows from the convergence of ρn to ρ in L1loc(Rd). We thus conclude that for vC1c(Ω),

    Gρn(v)Gρ(v)strongly in L2.

    Thus from the above integration by parts formula we have that for any vC1c(Ω)

    limnΩDρn(un)v(x)dx=limnΩGρn(v)(x)un(x)dx=ΩGρ(v)(x)u(x)dx=ΩDρ(u)v(x)dx.

    The last inequality is possible because uSρ,2(Ω). Now for any vL2(Ω), let us choose vmC1c(Ω) such that vmv strongly in L2(Ω). Then we have for each n,m that

    ΩDρn(un)v(x)dx=ΩDρn(un)vm(x)dx+Rn,m

    where

    |Rn,m|=|ΩDρn(un)(v(x)vm(x))dx|Dρn(un)L2(Ω)vmvL2(Ω).

    Therefore using the fact that Dρn(un)L2(Ω) is uniformly bounded in n, we have that limmsupnN|Rn,m|=0 and so we have

    lim infnΩDρn(un)v(x)dx=limnΩDρn(un)vm(x)dx+lim infnRn,m=ΩDρ(u)vm(x)dx+lim infnRn,m.

    We now take m and use the fact that Dρ(u)L2(Ω) to complete the proof the lemma.

    Lemma 2.2. Suppose that unu strongly in L2(Ω;Rd), uSρ,2(Ω), and that supnNWρn(un)C<. Then it holds that

    ΩΩρ(xx)(D(u)(x,x)1dDρ(u)(x))2dxdxlim infnΩΩρn(xx)(D(un)(x,x)1dDρn(un)(x))2dxdx. (2.1)

    Proof. Let A⊂⊂Ω and φCc(B1(0)). For ϵ<dist(A,Ω), consider the sequence of functions φϵun and φϵDρn(un) defined for xA, where φϵ(z)=ϵdφ(z/ϵ) is standard mollifiers. Then since unu strongly in L2, for a fixed ϵ>0, we have as n,

    φϵunφϵuin C2(¯A;Rd) andφϵDρn(un)φϵDρ(u)strongly in L2(A). (2.2)

    The latter follows from Lemma 2.1 and the fact that the convolution is a compact operator. Using Jensen's inequality, we have that for each ϵ>0 small and n large

    AAρn(xx)(D(φϵun)(x,x)1dφϵDρn(un)(x))2dxdxAAρn(xx)(D(un)(x,x)1dDρn(un)(x))2dxdx. (2.3)

    The left hand side of (2.3) can be rewritten after change of variables as

    AAρn(xx)(D(φϵun)(x,x)1dφϵDρn(un)(x))2dxdx=Rdρn(z)AχA(x+z)(D(φϵun)(x,x+z)1dφϵDρn(un)(x))2dxdz.

    Using (2.2), the sequence of functions

    zAχA(x+z)(D(φϵun)(x,x+z)1dφϵDρn(un)(x))2dx 

    converges in L(Ω) as n to

    zAχA(x+z)(D(φϵu)(x,x+z)1dφϵDρ(u)(x))2dx

    where we use the simple inequality |a2b2|||a|+|b|||ab| and the assumption that uSρ,2(Ω). Using the convergence of ρn to ρ in L1loc(Rd) and taking the limit in (2.3) we conclude that for each ϵ>0,

    AAρ(xx)(D(φϵu)(x,x)1dφϵDρ(u)(x))2dxdxlim infnAAρn(xx)(D(un)(x,x)1dDρn(un)(x))2dxdx.

    Now inequality (2.1) follows after applying first Fatou's lemma in ϵ and noting that A⊂⊂Ω was arbitrary.

    Let us state some elementary inequalities that relate the energy Wρ(u) and its integrand with that of the seminorm |u|Sρ,2. The proof follows from direct calculations and uses a simple application of Hölder's inequality.

    Lemma 2.3. For a given ρL1loc(Rd) and Ω bounded such that for uSρ,2(Ω) and xΩ we have

    Dρ(u)2(x)ρL1(BR(0))Ωρ(xy)|D(u)(x,y)|2dy,Ωρ(yx)(D(u)(x,y)1dDρ(u)(x))2dyC(d,ρL1(BR(0)))Ωρ(xy)|D(u)(x,y)|2dy.

    Moreover, we have positive constants C1 and C2, depending on ρ,Ω, and d, such that for all uSρ,2(Ω)

    C1|u|2Sρ,2(Ω)Wρ(u)C2|u|2Sρ,2(Ω).

    Proof of Theorem 1.2. The proof has two parts: the demonstration of the Γ-convergence of the energy functionals and the proof of the convergence of minimizers. For the first part, following the definition of Γ-convergence, we prove the two items in Definition 1.1.

    Item a) Suppose that unu strongly in L2. We will show that

    ¯Eρ(u)lim infn¯En(un).

    Since ΩfundxΩfudx as n, we only need to show that

    Wρ(u)lim infnWρn(un).

    To that end, we will assume without loss of generality that lim infnWρn(un)<, and so (up to a subsequence) supnNWρn(un)C<. Then we have that {Dρn(un)} is uniformly bounded in L2(Ω) and {|un|Sρn,2} is uniformly bounded as well, by Lemma 2.3. To prove the desired inequality it suffices to show that

    Ω(Dρ(u))2dxlim infnΩ(Dρn(un))2dx (2.4)

    and

    ΩΩρ(xx)(D(u)(x,x)1dDρ(u)(x))2dxdxlim infnΩΩρn(xx)(D(un)(x,x)1dDρn(un)(x))2dxdx. (2.5)

    To show (2.4), using the weak lower semicontinuity of norm, it suffices to show that Dρn(un)Dρ(u) weakly in L2(Ω). But this is proved in Lemma 2.1 after noting the above assumption.

    Inequality (2.5) will follow from Lemma 2.2 if we show uSρ,2(Ω). But under the assumptions on the sequence un, the conclusion uSρ,2(Ω) follows from Theorem 1.4 that will be proved in the coming sections.

    Item b) For a given uL2(Ω), we take the recovery sequence to be un=u. Now if uL2(Ω)Sρ,2(Ω), then by definition ˉE(u)= and necessarily lim infnEρn(u)=. Otherwise, up to a subsequence (nor renamed) supnEρn(u)< and

    [u]2Sρn,2C(Eρn(u)+uL2)C+uL2,

    where we used Lemma 2.3. Then by passing to the limit and using Fatou's lemma, we have [u]2Sρ,2<, that is, uSρ,2(Ω), which is a contradiction. In the event that uSρ,2(Ω), we may use (1.8) to get the pointwise convergence and Lemma 2.3 to get appropriate bounds of the integrand of Wρn(u) to apply the Dominated Convergence Theorem and conclude that lim infnWρn(u)=Wρ(u), from which Item b) follows.

    We next prove the second part of the theorem, the convergence of minimizers. To apply [21, Theorem 7.8 and Corollary 7.202], we need to prove the equicoercvity of the functionals restricted to VSρn,2. That is, for unVSρn,2 such that supn1Eρn(un)<, we need to show that the sequence {un} is precompact in L2(Ω;Rd). To that end, first a positive constant C>0 and for all n1

    [un]2Sρn,2C(Eρn(un)+unL2)C+unL2.

    Using the uniform Poincaré-Korn inequality, Theorem 1.5, for all large n we have that unL2C[un]Sρn,2 and as a consequence

    [un]2Sρn,2C(1+[un]Sρn,2)for all n large.

    It then follows that [un]Sρn,2 is uniformly bounded and therefore, by the uniform Poincaré-Korn inequality, unL2(Ω) is bounded as well. We now use the compactness result, Theorem 1.4, to conclude that {un} is precompact in L2(Ω;R2) with limit point u in Sρ,2(Ω)V. We may now apply [21, Theorem 7.8 and Corollary 7.202] to state that u is a minimizer of Eρ over Sρ,2(Ω)V.

    The proof of the Lp compactness result, Theorem 1.3, will be carried out in two steps. We establish first compactness in Lploc topology followed by proving a boundary estimate that controls growth near the boundary of the domain. The Lploc compactness will be proved in this section under a weaker assumption on the kernel. In fact Lploc compactness will be stated and proved for a broader class of kernels that include kernels of the type ˜ρ(ξ)χBΛ1(ξ) where ˜ρ satisfies (1.10) and (1.11), where BΛ1={xB1:x/|x|Λ} is a conic region spanned by a given a nontrivial spherical cap \Lambda \subset\mathbb{S}^{d-1} . To make this and the condition of the theorem precise, we begin identifying the kernel \rho by the representative

    {\rho}( {\bf x}) = \left\{\begin{aligned} &\lim\limits_{h\to 0}\rlap{-} \smallint _{B_{h}( {\bf x})} \rho({ \boldsymbol \xi})d{{ \boldsymbol \xi}}, \quad \text{if}\ {\bf x} \ \text{is a Lebesgue point}, \\ & \infty, \quad \text{otherwise}. \end{aligned}\right.

    For \theta_{0}\in (0, 1) and {\bf v}\in \mathbb{S}^{d-1} , let us define

    \rho_{\theta_{0}}(r{\bf v}) = \inf\limits_{\theta\in [\theta_{0}, 1]}\rho(\theta r{\bf v})\theta^{-p}.

    It is clear that for a given {\bf v}\in \mathbb{S}^{d-1} , \rho_{\theta_{0}}(r{\bf v}) \leq \rho(\theta r{\bf v})\theta^{-p} for any \theta\in [\theta_{0}, 1] and r\in (0, \infty) . In particular, this implies \rho_{\theta_{0}}({ \boldsymbol \xi}) \leq \rho({ \boldsymbol \xi}) for any { \boldsymbol \xi} , with the equality holds if \rho is radial and |{ \boldsymbol \xi}|^{-p}\rho({ \boldsymbol \xi}) is nonincreasing in |{ \boldsymbol \xi}| .

    We now make a main assumption on \rho that

    \begin{equation} \begin{aligned} \quad\quad & \exists\, \theta_{0}\in (0, 1) , \Lambda \subset \mathbb{S}^{d-1} \ {\rm{and}} \ {\bf v}_0\in \Lambda\ {\rm{such}}\ {\rm{that}} \ \mathcal{H}^{d-1}(\Lambda) > 0 , \\ & \rho_{\theta_{0}}(r{\bf v}) = \rho_{\theta_{0}}(r{\bf v}_0) , \forall\, (r, {\bf v}) \in (0, \infty)\times\Lambda , \, {\rm{and}} \; \lim\limits_{\delta \to 0}\frac{\delta^{p}}{ \int_{0}^{\delta} \rho_{\theta_{0}}(r{\bf v}_0) r^{d-1} dr} = 0. \end{aligned} \end{equation} (3.1)

    Assumption (3.1) says that, on a conic region with apex at the origin, the kernel \rho is above a nonnegative function with appropriate singular growth near the origin. Note that on one hand, it is not difficult to see if \rho \in L^{1}_{loc}(\mathbb{R}^{d}) is a nonnegative function that satisfies (1.10) and (1.11), then it also satisfies (3.1). On the other hand, if \tilde{\rho} satisfies (1.10) and (1.11), then given a nontrivial spherical cap \Lambda and conic region B_{1}^{\Lambda} = \{ {\bf x}\in B_{1}: { {\bf x}/|{ {\bf x}}|}\in \Lambda\} , the kernel \rho({ \boldsymbol \xi}) = \tilde{\rho} ({ \boldsymbol \xi})\chi_{B_{1}^{\Lambda}}({ \boldsymbol \xi}) satisfies (3.1) (with \theta_0 being any number in (0, 1) and {\bf v}_0 representing any vector in \Lambda ) but not necessarily (1.10) and (1.11). For kernels of this form, we need the formulation in (3.1) to carry out the proof of L^{p}_{loc} compactness. We should also note that one can construct other \rho that are not of the above form that satisfy (3.1), see [6, Eq (17)].

    Theorem 3.1 ( L^{p}_{loc} compactness). Suppose that 1\leq p < \infty . Let \rho \in L^{1}(\mathbb{R}^{d}) be a nonnegative function satisfying (3.1). Suppose also that \{{\bf u}_{n}\} is a sequence of vector fields that is bounded in \mathcal{S}_{\rho, p}(\mathbb{R}^{d}) . Then for any D\subset \mathbb{R}^{d} open and bounded, the sequence \{{\bf u}_{n}|_{D}\} is precompact in L^{p}(D; \mathbb{R}^{d}) .

    We should mention that although the focus is different, operators that use non-symmetric kernels like those satisfying the condition (3.1) have been studied in connection with semi-Dirichlet forms and the processes they generate, see [2,16] for more discussions. In particular, most of the examples of kernels listed in [16, Section 6] satisfy condition (3.1).

    We begin with the following lemma whose proof can be carried out following the argument used in [28]. Let {\bf u}\in L^{p}(\mathbb{R}^{d}; \mathbb{R}^{d}) be given, we introduce the function F_{p}[{\bf u}]:\mathbb{R}^{d}\to [0, \infty) defined by

    F_{p}[{\bf u}]({\bf h}) = \int_{\mathbb{R}^{d}}\left|({\bf u}( {\bf x} + {\bf h}) - {\bf u}( {\bf x}))\cdot \frac{{\bf h}}{|{\bf h}|}\right|^{p}d{ {\bf x}}, \quad \text{for}\ {\bf h}\in \mathbb{R}^{d} .

    Lemma 3.2. Suppose that \theta_{0} is given as in (3.1). There exists a constant C = C(\theta_{0}, p) > 0 such that for any \delta > 0 , and {\bf v}\in \mathbb{S}^{d-1}

    \begin{aligned} F_{p}[{\bf u}](t{\bf v}) &\leq C \frac{\delta^{p}}{ \int_{0}^{\delta} \rho_{\theta_{0}} (s{\bf v})s^{d-1} ds} \int_{0}^{\infty}\rho(h{\bf v}) h^{d-1}\frac{F_{p}[{\bf u}](h{\bf v})}{{h^{p}}}dh, \end{aligned}

    for any 0 < t < \delta and any {\bf u}\in L^{p}(\mathbb{R}^{d}, \mathbb{R}^{d}) .

    Proof. For any {\bf v}\in \mathbb{S}^{d-1} and t\in \mathbb{R} , we may rewrite the function F_{p} as

    F_{p}[{\bf u}](t{\bf v}) = \int_{\mathbb{R}^{d}}|({\bf u}( {\bf x} + t{\bf v}) - {\bf u}( {\bf x}))\cdot {\bf v}|^{p}d{ {\bf x}}.

    It follows from [28, Lemma 3.1] that given 0 < s < t , there exist C_{p} and \theta = \frac{t}{s} - k\in (0, 1) ( k an integer) such that

    \frac{F_{p}[{\bf u}](t{\bf v})}{ t^{p}} \leq C_{p}\left\{ {F_{p}[{\bf u}](s{\bf v})\over s^{p}} + {F_{p}[{\bf u}](\theta s{\bf v})\over t^{p}}\right\}.

    We also have that for a given l_{0} \in \mathbb{N} ,

    F_{p}[{\bf u}](\theta s{\bf v}) \leq l_{0}^{p} F_{p}[{\bf u}]\left(\frac{\theta s}{ l_{0}} {\bf v}\right) \leq {2^{(p-1)}}l^{p}_{0}\left\{ F_{p}[{\bf u}](s{\bf v}) + F_{p}[{\bf u}]\left(s - \frac{s\theta}{{l_{0}}}{\bf v}\right)\right\}.

    Combining the above we have that for any l_{0} , there exists a constant C = C(p, l_{0}) such that

    \begin{equation} \frac{F_{p}[{\bf u}](t{\bf v})}{ t^{p}} \leq C(p, l_{0})\left\{ {F_{p}[{\bf u}](s{\bf v})\over s^{p}} + {F_{p}[{\bf u}](\tilde{\theta} s{\bf v})\over t^{p}}\right\}, \quad \text{where}\ \tilde{\theta} = 1- {\theta\over l_{0}} . \end{equation} (3.2)

    Now let us take \theta_{0} as given in (3.1) and choose l_{0} large that {1\over l_{0}} < 1-\theta_{0} . It follows that \theta_{0} < \tilde{\theta} \leq 1 . Then for any \delta > 0 , and any 0 < s < \delta \leq \tau , by multiplying both sides of inequality (3.2) by \rho_{\theta_{0}} ({\bf v}s) and integrating from 0 to \delta , we obtain

    \begin{aligned} \int_{0}^{\delta} \rho_{\theta_{0}}(s{\bf v}) s^{d-1}ds {F_{p}[{\bf u}](\tau{\bf v})\over \tau^{p}} \leq C(p, l_{0})\left\{ \int_{0}^{\delta} \rho_{\theta_{0}}(s{\bf v}) s^{d-1}{F_{p}[{\bf u}](s{\bf v})\over s^{p}} ds \right. \left. + \int_{0}^{\delta} \rho_{\theta_{0}}(s{\bf v}) s^{d-1}{F_{p}[{\bf u}](\tilde{\theta} s{\bf v})\over \tau^{p}}ds \right\}. \end{aligned}

    Let us estimate the second integral in the above:

    I = {1\over \tau^{p}}\int_{0}^{\delta} \rho_{\theta_{0}}(s{\bf v}) s^{d-1}{F_{p}[{\bf u}](\tilde{\theta} s{\bf v})}ds.

    We first note that using the definition of \rho_{\theta_{0}} and since \delta \leq \tau , we have

    I\leq {1 \over {\theta_{0}^{d-1}}} \int_{0}^{\tau} \rho(\tilde{\theta}s{\bf v}) (\tilde{\theta} s)^{d-1}{F_{p}[{\bf u}](\tilde{\theta} s{\bf v})\over {(\tilde{\theta} s)^{p}}}ds.

    The intention is to change variables h = \tilde{\theta} s . However, note that \tilde{\theta} is a function of s , and by definition

    \tilde{\theta}s = \left(\frac{k}{ l_{0}} + 1\right)s - \frac{\tau}{ l_{0}}\quad \text{for}\ k \leq \frac{\tau}{s} < k + 1 .

    It then follows by a change of variables that

    \begin{aligned} I &\leq \frac{1} {{\theta_{0}^{d-1}}} \sum\limits_{k = 1}^{\infty}\int_{\frac{\tau}{ (k+1)}}^{\frac{\tau}{ k}} \rho(\tilde{\theta}s{\bf v}) (\tilde{\theta}s)^{d-1}\frac{F_{p}[{\bf u}](\tilde{\theta} s{\bf v})}{ {(\tilde{\theta} s)^{p}}}ds\\ & = \frac{1}{{\theta_{0}^{d-1}}} \sum\limits_{k = 1}^{\infty}\int_{{\tau(1 - \frac{1}{ l_{0}})\over (k+1)}}^{\frac{\tau}{ k}} \rho(h{\bf v}) h^{d-1}{F_{p}[{\bf u}](h{\bf v})\over {h^{p}}}{dh\over {k\over l_{0}} + 1}\\ &\leq C \int_{0}^{\infty}\rho(h{\bf v}) h^{d-1}\frac{F_{p}[{\bf u}](h{\bf v})}{h^{p}}dh, \end{aligned}

    where in the last estimate integrals over overlapping domains were counted at most a finite number of times. Combining the above estimates we have shown that there exists a constant C such that for any {\bf v} \in \mathbb{S}^{d-1} , \delta > 0 and \tau\geq \delta

    \left(\int_{0}^{\delta} \rho_{\theta_{0}} (s{\bf v})s^{d-1} ds\right) \frac{F_{p}[{\bf u}](\tau{\bf v})}{ \tau^{p}} \leq C \int_{0}^{\infty}\rho(h{\bf v}) h^{d-1}\frac{F_{p}[{\bf u}](h{\bf v})}{ {h^{p}}}dh.

    Rewriting the above and restricting {\bf v} \in \Lambda we have that

    F_{p}[{\bf u}](\tau{\bf v}) \leq C \frac{\tau^{p}}{ \int_{0}^{\delta} \rho_{\theta_{0}} (s{\bf v})s^{d-1} ds} \int_{0}^{\infty}\rho(h{\bf v}) h^{d-1}\frac{F_{p}[{\bf u}](h{\bf v})}{h^{p}}dh.

    Now let 0 < t < \delta and applying the above inequality for \tau = \delta and \tau = t + \delta , we obtain

    \begin{aligned} F_{p}[{\bf u}](t{\bf v}) & = F_{p}[{\bf u}]((t + \delta){\bf v} -\delta{\bf v})\\ &\leq 2^{p-1} \left\{F_{p}[{\bf u}]((t + \delta){\bf v}) + F_{p}[{\bf u}](\delta{\bf v})\right\}\\ &\leq C \frac{\delta^{p}}{ \int_{0}^{\delta} \rho_{\theta_{0}} (s{\bf v})s^{d-1} ds} \int_{0}^{\infty}\rho(h{\bf v}) h^{d-1}\frac{F_{p}[{\bf u}](h{\bf v})}{ {h^{p}}}dh. \end{aligned}

    This completes the proof.

    Lemma 3.3. Suppose that \rho\in L^{1}_{loc}(\mathbb{R}^{d}) and there exists a spherical cap \Lambda\subset \mathbb{S}^{d-1} and a vector {\bf v}_{0}\in\Lambda such that the function \rho(r{\bf v}) = \rho(r{\bf v}_{0}) = \tilde{\rho}(r) , for all {\bf v}\in \Lambda and r\mapsto r^{-p}\tilde{\rho}(r) is nonincreasing. Then there exists a constant C = C(d, p, \Lambda) such that for any \delta > 0 , and {\bf v}\in \Lambda ,

    \begin{aligned} F_{p}[{\bf u}](t{\bf v}) &\leq C \frac{\delta^{p}}{ \int_{0}^{\delta} \tilde{\rho} (s)s^{d-1} ds} \int_{0}^{\infty}{\rho}(h{\bf v}) h^{d-1}\frac{F_{p}[{\bf u}](h{\bf v})}{{h^{p}}}dh, \end{aligned}

    for any 0 < t < \delta and any {\bf u}\in L^{p}(\mathbb{R}^{d}, \mathbb{R}^{d}) .

    Proof. It suffices to note that for \rho\in L^{1}_{loc}(\mathbb{R}^{d}) that satisfies the conditions in the statement of the proposition, we have that for any \theta_{0} \in (0, 1) , and any {\bf v}\in \Lambda ,

    \rho_{\theta_{0}} (r{\bf v}) = r^{p} \inf\limits_{\theta\in [\theta_{0}, 1]} \rho(\theta r{\bf v})(\theta r)^{-p} = \rho(r{\bf v}) = \rho(r{\bf v}_{0}) = \tilde{\rho}(r).

    We may then repeat the argument in the proof of Lemma 3.2.

    Before proving one of the main results, we make an elementary observation.

    Lemma 3.4. Let 1\leq p < \infty. Given a spherical cap \Lambda with aperture \theta , there exists a positive constant c_{0}, depending only on d, \theta and p , such that

    \begin{equation*} \inf\limits_{{\bf w}\in \mathbb{S}^{d-1}}\int_{\Lambda\cap \mathbb{S}^{d-1}} |{\bf w}\cdot {\bf s}|^{p}d\sigma({\bf s}) \geq c_{0} > 0. \end{equation*}

    The above lemma follows from the fact that the map

    {\bf w}\mapsto \int_{\Lambda\cap\mathbb{S}^{d-1}} |{\bf w}\cdot {\bf s}|^{p}d\sigma({\bf s})

    is continuous on the compact set \mathbb{S}^{d-1}, and is positive, for otherwise the portion of the unit sphere \Lambda will be orthogonal to a fixed vector which is not possible since \mathcal{H}^{d-1}(\Lambda) > 0 .

    From the assumption we have

    \begin{equation} \sup\limits_{n\geq 1} \|{\bf u}_{n}\|_{L^{p}}^{p} + \sup\limits_{n\geq 1}\int_{\mathbb{R}^{d}}\int_{\mathbb{R}^{d}}\rho( {\bf x}'- {\bf x})\left| {\mathscr D}({\bf u}_n)( {\bf x}, {\bf x}')\right|^{p}d {\bf x}'d {\bf x} < \infty. \end{equation} (3.3)

    We will use the compactness criterion in [20, Lemma 5.4], which is a variant of the well-known Riesz-Fréchet-Kolmogorov compactness criterion [7, Chapter Ⅳ.27]. Let \Lambda be as given in (3.1). For \delta > 0 , let us introduce the matrix Q = (q_{ij}) , where

    q_{ij} = \int_{\Lambda} s_{i}s_{j} d \mathcal{H}^{d-1}({\bf s}).

    The symmetric matrix \mathbb{Q} is invertible. Indeed, the smallest eigenvalue is given by

    \lambda_{min} = \min\limits_{| {\bf x}| = 1} \langle \mathbb{Q} {\bf x}, {\bf x} \rangle = \min\limits_{| {\bf x}| = 1} \int_{\Lambda} | {\bf x}\cdot {\bf s}|^{2} d \mathcal{H}^{d-1}({\bf s})

    which we know is positive by Lemma 3.4. We define the following matrix functions

    \begin{equation*} \mathbb{P} ({\bf z}) = d\mathbb{Q}^{-1}\frac{{\bf z}\otimes {\bf z}}{|{\bf z}|^{2}}\chi_{B^{\Lambda}_{1}}({\bf z}), \quad \quad \mathbb{P}^{\delta}({\bf z}) = \delta^{-d}\mathbb{P} \left(\frac{{\bf z}}{\delta}\right) \end{equation*}

    where B_{1}^{\Lambda} = \{ {\bf x}\in B_{1}: { {\bf x}/|{ {\bf x}}|}\in \Lambda\} , as defined before. Then for any \delta > 0 ,

    \int_{\mathbb{R}^{d}} \mathbb{P}^{\delta}({\bf z}) d{\bf z} = \mathbb{I}.

    To prove the theorem, using [20, Lemma 5.4], it suffices to prove that

    \begin{equation} \lim\limits_{\delta\to 0}\limsup\limits_{n\to \infty}\|{\bf u}_{n} - \mathbb{P}^{\delta}*{\bf u}_{n}\|_{L^{p}(\mathbb{R}^{d})} = 0. \end{equation} (3.4)

    We show next that the inequality (3.3) and condition (3.1) imply (3.4). To see this, we begin by introducing the notation B_{\delta}^{\Lambda} = \{ {\bf x}\in B_{\delta}({ \bf 0}): { {\bf x}/|{ {\bf x}}|}\in \Lambda\} and applying Jensen's inequality to get

    \begin{equation} \begin{aligned} \int_{\mathbb{R}^{d}} |{\bf u}_{n}( {\bf x}) - \mathbb{P}^{\delta}*{\bf u}_{n}( {\bf x})|^{p}d {\bf x} &\leq \int_{\mathbb{R}^{d}}\left|\int_{\mathbb{R}^{d}}\mathbb{P}^{\delta}( {\bf y} - {\bf x})({\bf u}_{n}( {\bf y}) - {\bf u}_{n}( {\bf x})) d {\bf y}\right|^{p}d {\bf x}\\ &\leq \int_{\mathbb{R}^{d}}\left||\Lambda|\mathbb{Q}^{-1}\rlap{-} \smallint _{B^{\Lambda}_{\delta}( {\bf x})}\frac{( {\bf y}- {\bf x})}{| {\bf y}- {\bf x}|}\cdot({\bf u}_{n}( {\bf y}) - {\bf u}_{n}( {\bf x})) \frac{( {\bf y}- {\bf x})}{| {\bf y}- {\bf x}|}d {\bf y}\right|^{p}d {\bf x}\\ &\leq |\Lambda|^{p}\|\mathbb{Q}^{-1}\|^{p}\int_{\mathbb{R}^{d}}\left|\rlap{-} \smallint _{B^{\Lambda}_{\delta}( {\bf x})}\frac{( {\bf y}- {\bf x})}{| {\bf y}- {\bf x}|}\cdot({\bf u}_{n}( {\bf y}) - {\bf u}_{n}( {\bf x})) \frac{( {\bf y}- {\bf x})}{| {\bf y}- {\bf x}|}d {\bf y}\right|^{p}d {\bf x}\\ &\leq\frac{ |\Lambda|^{p}\|\mathbb{Q}^{-1}\|^{p}}{|B^{\Lambda}_{\delta}|}\int_{0}^{\delta}\int_{\Lambda}\tau^{d-1}{F}_{p}[{\bf u}_{n}](\tau{\bf v}) d \mathcal{H}^{d-1}({\bf v})d\tau\\ &\leq \frac{C(d, p)}{|B^{\Lambda}_{\delta}|} \int_{0}^{\delta}\int_{\Lambda}\tau^{d-1}{F}_{p}[{\bf u}_{n}](\tau{\bf v}) d \mathcal{H}^{d-1}({\bf v})d\tau \end{aligned} \end{equation} (3.5)

    where as defined previously

    {F}_{p}[{\bf u}_{n}](\tau {\bf v}) = \int_{\mathbb{R}^{d}} \left|{\bf v}\cdot({\bf u}_{n}( {\bf x}+\tau {\bf {\bf v}}) - {\bf u}_{n}( {\bf x}))\right|^{p}d {\bf x}.

    Moreover, the fact that |\Lambda|^{p}\|\mathbb{Q}^{-1}\|^{p} \leq C(d, p, \Lambda) for any \delta > 0 is also used. We can now apply Lemma 3.2 and use the condition (3.1) to obtain that

    \begin{aligned} & \frac{ C(d, p, \lambda)}{|B^{\Lambda}_{\delta}|}\int_{0}^{\delta}\int_{\Lambda}\tau^{d-1}{F}_{p}[{\bf u}_{n}](\tau{\bf v}) d \mathcal{H}^{d-1}({\bf v})d\tau\\ &\quad \leq \frac{ C(d, p, \Lambda)}{|B^{\Lambda}_{\delta}|}\int_{0}^{\delta}\tau^{d-1} d \tau\int_{\Lambda} \left( { \frac{\delta^{p}}{ \int_{0}^{\delta} \rho_{\theta_{0}} (s{\bf v}_{0})s^{d-1} ds } \int_{0}^{\infty}\rho(h{\bf v}) h^{d-1}\frac{F_{p}[{\bf u}_{n}](h{\bf v})}{{h^{p}}}dh} \right) d\mathcal{H}^{d-1}({\bf v})\\ &\quad \leq C(d, p, \Lambda) \frac{\delta^{p}}{ \int_{0}^{\delta} \rho_{\theta_{0}} (s{\bf v}_{0})s^{d-1} ds} |{\bf u}_{n}|_{\mathcal{S}_{\rho, p}(\mathbb{R}^{d})}. \end{aligned}

    Therefore from the boundedness assumption (3.3) we have,

    \begin{aligned} \int_{\mathbb{R}^{d}}&|{\bf u}_{n}( {\bf x}) - \mathbb{P}^{\delta}*{\bf u}_{n}( {\bf x})|^{p}d {\bf x} \leq C(p, d, \Lambda) \frac{\delta^{p}}{ \int_{0}^{\delta} \rho_{\theta_{0}} (s{\bf v}_{0})s^{d-1} ds}. \end{aligned}

    Equation (3.4) now follows from condition (3.1) after letting \delta\to 0 . That completes the proof.

    A corollary of the compactness result, Theorem 3.1, is the following result that uses a criterion involving a sequence of kernels. The effort made in the proof above was to show the theorem for kernel \rho satisfying (3.1), but the proposition below limits to those satisfying (1.10) and (1.11).

    Proposition 3.5. Let \rho \in L^{1}_{loc} satisfy (1.10) and (1.11). Let \rho_{n} be a sequence of radial functions satisfying (1.10) and that \rho_{n}\rightharpoonup \rho weakly in L^{1} as n\to \infty . If

    \sup\limits_{n\geq 1}\{ \|{\bf u_{n}}\|_{L^{p}(\mathbb{R}^{d})} + |{\bf u}_{n}|_{\mathcal{S}_{\rho_{n}, p}}\} < \infty

    then \{{\bf u}_{n}\} is precompact in L^{p}_{loc}(\mathbb{R}^{d}; \mathbb{R}^{d}) . Moreover, if A\subset \mathbb{R}^{d} is a compact subset, the limit point of the sequence restricting to A is in \mathcal{S}_{\rho, p}(A).

    Proof. Using Lemma 3.3 applied to each \rho_{n} , we can repeat the argument in the proof of Theorem 3.1 to obtain

    \int_{\mathbb{R}^{d}}|{\bf u}_{n}( {\bf x}) - \mathbb{P}^{\delta}*{\bf u}_{n}( {\bf x})|^{p}d {\bf x} \leq C(p, d) \frac{\delta^{p}}{ \int_{0}^{\delta} \rho_{n}(r)r^{d-1} dr } \leq C(p, d) \frac{\delta^{p}}{ \int_{B_{\delta}} \rho_{n}({ \boldsymbol \xi}) d{ \boldsymbol \xi} } \, .

    Now since \rho_{n}\rightharpoonup \rho , weakly in L^{1} as n\to \infty , for a fixed \delta > 0 , it follows that

    \limsup\limits_{n\to \infty} \int_{\mathbb{R}^{d}}|{\bf u}_{n}( {\bf x}) - \mathbb{P}^{\delta}*{\bf u}_{n}( {\bf x})|^{p}d {\bf x} \leq C(p, d) \frac{\delta^{p}}{ \int_{B_{\delta}} \rho({ \boldsymbol \xi}) d{ \boldsymbol \xi} } \, .

    We now let \delta \to 0 , and use the assumption (1.11) to obtain

    \lim\limits_{\delta\to 0}\limsup\limits_{n\to \infty} \int_{\mathbb{R}^{d}}|{\bf u}_{n}( {\bf x}) - \mathbb{P}^{\delta}*{\bf u}_{n}( {\bf x})|^{p}d {\bf x} = 0,

    from which the compactness in the L^{p}_{loc} topology follows.

    We next prove the final conclusion of the proposition. To that end, let A \subset \mathbb{R}^{d} be a compact subset. For \phi\in C_{c}^{\infty}(B_{1}) , we consider the convoluted sequence of function \phi_{\epsilon} *{\bf u}_{n} , where \phi_{\epsilon}({\bf z}) = \epsilon^{-d} \phi({\bf z}/\epsilon) is the standard mollifier. Since {\bf u}_{n} \to {\bf u} strongly in L^{p}(A; \mathbb{R}^{d}) for a fixed \epsilon > 0 , we have as n\to\infty ,

    \begin{equation} \phi_{\epsilon} *{\bf u}_{n} \to \phi_{\epsilon} *{\bf u} \quad \text{in $C^{2}(A;\mathbb{R}^{d})$}. \end{equation} (3.6)

    Using Jensen's inequality, we obtain that for any \epsilon > 0 , and n large,

    \begin{aligned} \int_{A}&\int_{A} \rho_{n}( {\bf y}- {\bf x})\left|\frac{(\phi_{\epsilon} *{\bf u}_{n} ( {\bf y}) - \phi_{\epsilon} *{\bf u}_{n} ( {\bf x})) \cdot ( {\bf y} - {\bf x})}{ | {\bf y} - {\bf x}|^{2}}\right|^{p} d {\bf y} d {\bf x}\\ &\leq \int_{\mathbb{R}^{d}}\int_{\mathbb{R}^{d}} \rho_{n}( {\bf y}- {\bf x}) \left| \frac{ ({\bf u}_{n} ( {\bf y}) - {\bf u}_{n} ( {\bf x})) \cdot ( {\bf y} - {\bf x})}{| {\bf y} - {\bf x}|^{2}}\right|^{p}d {\bf y} d {\bf x}. \end{aligned}

    Taking the limit in n for fixed \epsilon , we obtain for any A compact that

    \int_{A}\int_{A} \rho( {\bf y}- {\bf x}) \left| \frac{(\phi_{\epsilon} *{\bf u} ( {\bf y}) - \phi_{\epsilon} *{\bf u} ( {\bf x})) \cdot ( {\bf y} - {\bf x})}{| {\bf y} - {\bf x}|^{2}}\right|^{p} d {\bf y} d {\bf x} \leq \sup\limits_{n\geq 1} {|{\bf u}_{n}|_{\mathcal{S}_{\rho_{n}, p}}^{p}} < \infty

    where we have used (3.6) and the fact that \rho_{n} converges weakly to \rho in L^{1} . Finally, let \epsilon \to 0 and use Fatou's lemma (since \phi_{\epsilon} *{\bf u}\to {\bf u} almost everywhere) to obtain that for any compact set A ,

    \int_{A}\int_{A} \rho( {\bf y}- {\bf x}) \left| \frac{({\bf u} ( {\bf y}) - {\bf u} ( {\bf x})) \cdot ( {\bf y} - {\bf x})} {| {\bf y} - {\bf x}|^{2}}\right|^{p} d {\bf y} d {\bf x} \leq \sup\limits_{n\geq 1} {|{\bf u}_{n}|_{\mathcal{S}_{\rho_{n}, p}}^{p}} < \infty,

    hence completing the proof.

    In this section we prove Theorem 1.3. We follow the approach presented in [28]. The argument relies on controlling the L^{p} mass of each {{\bf u}_{n}} , \int_{\Omega} |{\bf u}_{n}| ^{p} d {\bf x} , near the boundary by using the bound on the seminorm to demonstrate that in the limit when n\to \infty there is no mass concentration or loss of mass at the boundary. This type of control has been done for the sequence of kernels that converge to the Dirac Delta measure in the sense of measures. We will do the same for a fixed locally integrable kernel \rho satisfying the condition (1.11).

    In order to control the behavior of functions near the boundary by the semi-norm |\cdot|_{\mathcal{S}_{p, \rho}} , we first present a few technical lemmas.

    Lemma 4.1. [28] Suppose that 1\leq p < \infty and that g\in L^{p}(0, \infty) . Then there exists a constant C = C(p) such that for any \delta > 0 and t\in (0, \delta)

    \int_{0}^{\delta} |g(x)|^{p} dx \leq C \delta^{p} \int_{0}^{2\delta} \frac{|g(x +t )-g(x)|^{p}}{t^{p}} dx + 2^{p-1}\int_{\delta}^{3\delta} |g(x)|^{p} dx.

    Proof. For a given t\in (0, \delta) , choose k to be the first positive integer such that kt > \delta . Observe that (k-1)t \leq \delta , and so kt \leq 2\delta . Now let us write

    \begin{aligned} |g(x)|^{p}&\leq 2^{p-1} (|g(x + kt) - g(x)|^{p} + |g(x + kt)|^{p} )\\ &\leq 2^{p-1} k^{p-1} \sum\limits_{j = 0}^{k-1}|g(x + jt + t) - g(x {+ jt})|^{p} + 2^{p-1} |g(x + kt)|^{p}. \end{aligned}

    We now integrate in x on both side over (0, \delta) to obtain that

    \begin{aligned} \int_{0}^{\delta}|g(x)|^{p} dx &\leq 2^{p-1} k^{p-1} \sum\limits_{j = 0}^{k-1}\int_{0}^{\delta}|g(x + jt + t) - g(x + jt)|^{p}dx + 2^{p-1} \int_{0}^{\delta}|g(x + kt)|^{p}dx\\ &\leq 2^{p-1} k^{p-1} \sum\limits_{j = 0}^{k-1}\int_{jt}^{\delta + jt}|g(x + t) - g(x)|^{p}dx + 2^{p-1} \int_{\delta}^{3\delta}|g(x)|^{p}dx\\ &\leq 2^{p-1} k^{p} \int_{0}^{2\delta }|g(x + t) - g(x)|^{p}dx + 2^{p-1} \int_{\delta}^{3\delta}|g(x)|^{p}dx\, . \end{aligned}

    Recalling that kt \leq 2\delta , we have that k^{p} \leq 2^{p}\delta^{p}/t^{p} and we finally obtain the conclusion of the lemma with C = 2^{2p-1} .

    The above lemma will be used on functions of type t\mapsto {\bf u}({\bf x} + t{\bf v})\cdot {\bf v} , for {\bf v}\in \mathbb{S}^{d-1} . Before doing so, we need to make some preparation first. Observe that since \Omega is a bounded open subset of \mathbb{R}^{d} with a Lipschitz boundary, there exist positive constants r_{0} and \kappa with the property that for each point { \boldsymbol \xi}\in \partial \Omega there corresponds a coordinate system ({\bf x}', x_{d}) with {\bf x}'\in \mathbb{R}^{d-1} and x_{d}\in\mathbb{R} and a Lipschitz continuous function \zeta:\mathbb{R}^{d-1}\to \mathbb{R} such that |\zeta({\bf x}')-\zeta({\bf y}')|\leq \kappa| {\bf x}'- {\bf y}'| ,

    \begin{equation*} \Omega\cap B({ \boldsymbol \xi}, 4r_{0}) = \{( {\bf x}', x_{d}): x_{d} > \zeta( {\bf x}')\}\cap B({ \boldsymbol \xi}, 4r_{0}), \end{equation*}

    and \partial\Omega\cap B({ \boldsymbol \xi}, 4r_{0}) = \{({\bf x}', x_{d}): x_{d} = \zeta({\bf x}')\}\cap B({ \boldsymbol \xi}, 4r_{0}). It is well known that a Lipschitz domain has a uniform interior cone \Sigma({ \boldsymbol \xi}, \theta) at every boundary point { \boldsymbol \xi} such that B({ \boldsymbol \xi}, 4\, r_{0})\cap \Sigma({ \boldsymbol \xi}, \theta)\subset \Omega. The uniform aperture \theta \in(0, \pi) of such cones depends on the Lipschitz constant \kappa of the local defining function \zeta, and does not depend on { \boldsymbol \xi}. It is not difficult either to see that for any r\in (0, 4r_{0}), if {\bf y} \in B_{r}({ \boldsymbol \xi}), then

    \begin{equation*} \text{dist}( {\bf y}, \partial \Omega) = \inf\{|{ {\bf y}-( {\bf x}', x_{d})}|: ( {\bf x}', x_{d})\in B_{3r}({ \boldsymbol \xi}), x_{d} = \zeta( {\bf x}')\}. \end{equation*}

    We now begin to work on local boundary estimates. To do that without loss of generality, see Figure 1 below, after translation and rotation (if necessary) we may assume that { \boldsymbol \xi} = { \bf 0} and

    \begin{equation*} \Omega\cap B({ \bf 0}, 4r_{0}) = \{( {\bf x}', x_{d}): x_{d} > \zeta( {\bf x}')\}\cap B({ \bf 0}, 4r_{0}), \end{equation*}
    Figure 1.  \Sigma is the cone with aperture \pi/4 (depicted by the blue lines). The Lipschitz graph \zeta remains outside the double cone with aperture \pi/2 - \arctan(1/2) (depicted by the red lines). The red dashed line has length r .

    where \zeta({ \bf 0}')=0, and |\zeta({\bf x}')-\zeta({\bf y}')|\leq \kappa| {\bf x}'- {\bf y}'| . We also assume that the Lipschitz constant \kappa = 1/2 and the uniform aperture \theta = {\pi}/{4}. As a consequence, \zeta({\bf x}')\leq | {\bf x}'|/2 for all {\bf x}'\in B_{4r_{0}}({ \bf 0}'). Given any 0 < r < r_{0}, we consider the graph of \zeta :

    \begin{equation*} \Gamma_{r}: = \{ {\bf x} = ( {\bf x}', \zeta( {\bf x}'))\in \mathbb{R}^{d}: {\bf x}'\in B_{r}({ \bf 0}')\}. \end{equation*}

    We denote the upper cone with aperture {\pi}/{4} by \Sigma and is given by

    \begin{equation*} \Sigma = \{ {\bf x} = ( {\bf x}', x_{d})\in \mathbb{R}^{d}: | {\bf x}'|\leq x_{d}\}. \end{equation*}

    Finally we define \Omega_{\tau} = \{ {\bf x}\in \Omega: \text{dist}({\bf x}, \partial\Omega) > \tau\} to be the set of points in \Omega at least r units away from the boundary. Based on the above discussion we have that for any r\in (0, r_{0}],

    \begin{equation} \Omega\cap B_{r/2}\subset \Gamma_{r} + (\Sigma \cap B_{r}) \subset \Omega\cap B_{3r}. \end{equation} (4.1)

    Indeed, let us pick {\bf x} = ({\bf x}', x_{d})\in \Omega\cap B_{r/2} . The point \boldsymbol{{ \boldsymbol \xi}} = {\bf x}- ({\bf x}', \zeta({\bf x}')) = ({ \bf 0}', x_{d}-\zeta({\bf x}'))\in \Sigma, since {0 < } x_{d}-\zeta({\bf x}'). Moreover, by the bound on the Lipschitz constant |\boldsymbol{{ \boldsymbol \xi}}| = |x_{d}-\zeta({\bf x}')| < r/2 + r/4 < r. On the other hand, for any

    {\bf x} = ( {\bf x}_{1}', \zeta( {\bf x}'_{1})) + ( {\bf x}'_{2}, (x_{2})_{d})\in \Gamma_{r} + ( {\Sigma} \cap B_{r}),

    we have

    \zeta( {\bf x}_{1}' + {\bf x}'_{2})-\zeta( {\bf x}_{1}') \leq {| {\bf x}'_{2}|}/{2} \leq {(x_{2})_{d}}/{2},

    showing that \zeta({\bf x}_{1}' + {\bf x}'_{2}) < \zeta({\bf x}_{1}') + (x_{2})_{d} and therefore {\bf x}\in \Omega . It easily follows that {\bf x}\in B_{3r}, as well.

    For any r\in (0, r_0) , and {\bf x}'\in B_{r} ({ \bf 0}') , and {\bf v} \in \Sigma\cap\mathbb{S}^{d-1} , \text{dist}(({\bf x}', \zeta({\bf x}')) +r {\bf v}, \partial\Omega)\geq r/\sqrt{10} . Indeed, \text{dist}(({\bf x}', \zeta({\bf x}')) +r {\bf v}, \partial\Omega) is larger than or equal to the length of the black dashed line, which is larger than or equal to r\sin(\pi/4-\arctan(1/2)) = r/\sqrt{10} .

    In this subsection we establish the near boundary estimate in the following lemma.

    Lemma 4.2. Suppose that \Omega\subset \mathbb{R}^{d} is a domain with Lipschitz boundary. Let 1\leq p < \infty . Then there exist positive constants C_{1}, C_{2} , r_{0} and \epsilon_{0}\in (0, 1) with the property that for any r\in (0, r_{0}), {\bf u}\in L^{p}(\Omega; \mathbb{R}^{d}), and any nonnegative and nonzero \rho\in L^{1}_{loc}(\mathbb{R}^{d}) that is radial, we have

    \begin{equation*} \int_{\Omega}|{\bf u}|^{p}d {\bf x} \leq C_{1}(r) \int_{\Omega_{\epsilon_{0}r}}|{\bf u}|^{p}d {\bf x} + {C_{2} \frac{r^{p}}{ \int_{B_{r}(0)}\rho({\bf h}) d{\bf h}}} \int_{\Omega}\int_{\Omega}\rho( {\bf x}- {\bf y})\left| {\mathscr D}({\bf u})( {\bf x}, {\bf y})\right|^{p} d {\bf x}\, d {\bf y}. \end{equation*}

    The constant C_{1} may depend on r but the other constants C_{2} and r_{0} depend only on d, p and the Lipschitz constant of \Omega . Here for any \tau > 0 , we define \Omega_{\tau} = \{ {\bf x}\in \Omega: \mathit{\text{dist}}({\bf x}, \partial\Omega) > \tau\} .

    Proof. Following the above discussion, let us pick \overline{ \boldsymbol \eta}\in \partial \Omega and assume without loss of generality that \overline{ \boldsymbol \eta} = { \bf 0}, the function \zeta that defines the boundary \partial \Omega has a Lipschitz constant not bigger than 1/2 and the aperture is \pi/4. Assume first that {\bf u} \in L^{p}(\Omega; \mathbb{R}^{d}) , and vanishes on \Omega_{r/\sqrt{10}} . Let us pick \boldsymbol{{ \boldsymbol \xi}} = ({\bf x}', \zeta({\bf x}')) such that | {\bf x}'| < r and {\bf v}\in \Sigma \cap \mathbb{S}^{d-1}. Let us introduce the function

    \begin{equation*} g_{\bf v}^{{ \boldsymbol \xi}}(t) = {\bf u}({ \boldsymbol \xi}+ t{\bf v})\cdot {\bf v}, \quad t\in (0, 3 r_{0}). \end{equation*}

    Then for all { \boldsymbol \xi}\in \Gamma_{r} and {\bf v}\in \Sigma\cap\mathbb{S}^{d-1} , { \boldsymbol \xi} + r{\bf v}\in \Omega_{r/\sqrt{10}} . It follows that, by assumption on the vector field {\bf u}, the function g_{\bf v}^{{ \boldsymbol { \boldsymbol \xi}}}(t) \in L^{p}(0, 2r) and g_{\bf v}^{ \boldsymbol { \boldsymbol \xi}}(t) = 0 for t\in (r, 2r). We then apply Lemma 4.1 to get a constant C_{p} > 0 such that for any t\in(0, r) ,

    \begin{equation*} \int_{0}^{r}|{\bf u}({ \boldsymbol { \boldsymbol \xi}}+ s{\bf v})\cdot {\bf v}|^{p}\, ds \leq C \, r^{p}\int_{0}^{r} \frac{|({\bf u}({ \boldsymbol { \boldsymbol \xi}}+ s{\bf v} +t{\bf v} )-{\bf u}({ \boldsymbol { \boldsymbol \xi}}+ s{\bf v}))\cdot {\bf v}|^{p}}{t^{p}}ds, \end{equation*}

    where we used the fact that u vanishes on \Omega_{r/\sqrt{10}} . Noting that \boldsymbol{{ \boldsymbol \xi}} = ({\bf x}', \zeta({\bf x}')) for some {\bf x}'\in B'_{r}\subset \mathbb{R}^{d-1}, we integrate first in the above estimate with respect to {\bf x}'\in B'_{r} to obtain that

    \begin{equation*} \begin{aligned} & \int_{B'_{r}}\int_{0}^{r}|{\bf u}({ \boldsymbol \xi}+ s{\bf v})\cdot {\bf v}|^{p}\, ds\, d {\bf x}' \\ &\qquad \leq C \, r^{p}\int_{B'_{r}}\int_{0}^{r} \frac{|({\bf u}({ \boldsymbol \xi}+ s{\bf v} +t{\bf v} )-{\bf u}({ \boldsymbol \xi}+ s{\bf v}))\cdot {\bf v}|^{p}}{t^{p}}ds\, d {\bf x}' . \end{aligned} \end{equation*}

    The next step involves a change of variable {\bf y} = ({\bf x}', \zeta({\bf x}')) + s{\bf v} . Define a mapping G: ({\bf x}', s) \mapsto ({\bf x}', \zeta({\bf x}')) + s{\bf v} . Then the Jacobian of the mapping is defined almost everywhere and is given by

    J_G = {\bf v} \cdot (-\nabla \zeta^T( {\bf x}'), 1).

    Notice that |J_G| is bounded from above and below by two constants since {\bf v}\in \Sigma\cap\mathbb{S}^{d-1} and the Lipschitz constant of \zeta is not bigger than 1/2 . Also notice that G(B_r'\times (0, 1)) \subset \Gamma_{r} + \Sigma\cap B_{r} . Therefore,

    \begin{equation*} \begin{aligned} \int_{B'_{r}}&\int_{0}^{r} \frac{|({\bf u}({ \boldsymbol \xi}+ s{\bf v} +t{\bf v} )-{\bf u}({ \boldsymbol \xi}+ s{\bf v}))\cdot {\bf v}|^{p}}{t^{p}}ds\, d {\bf x}' \\ &\leq C \int_{\Gamma_{r} + \Sigma\cap B_{r}} \frac{|({\bf u}( {\bf y} +t{\bf v} )-{\bf u}( {\bf y}))\cdot {\bf v}|^{p}}{t^{p}}d {\bf y}\\ &\leq C \int_{\Omega\cap B_{3r}} \frac{|({\bf u}( {\bf y} +t{\bf v} )-{\bf u}( {\bf y}))\cdot {\bf v}|^{p}}{t^{p}}d {\bf y}, \end{aligned} \end{equation*}

    where we have used (4.1) in the last step. By some straightforward calculations, one can also find that \Omega\cap B_{r/4} \subset G(B_r'\times (0, 1)) . Then

    \int_{\Omega\cap B_{\frac{r}{4}}}|{\bf u}( {\bf y})\cdot {\bf v}|^{p}d {\bf y} \leq \int_{G(B_r'\times (0, 1))}|{\bf u}( {\bf y})\cdot {\bf v}|^{p}d {\bf y} \leq C \int_{B'_{r}}\int_{0}^{r}|{\bf u}({ \boldsymbol \xi}+ s{\bf v})\cdot {\bf v}|^{p}\, ds\, d {\bf x}'.

    It then follows from the above calculations that that for all {\bf v}\in \Sigma\cap\mathbb{S}^{d-1} and all t\in (0, r),

    \begin{equation} \int_{\Omega\cap B_{ \frac{r}{4}}}|{\bf u}( {\bf y})\cdot {\bf v}|^{p}d {\bf y} \leq C \, r^p\int_{\Omega\cap B_{3r}} \frac{|({\bf u}( {\bf y} +t{\bf v} )-{\bf u}( {\bf y}))\cdot {\bf v}|^{p}}{t^{p}}d {\bf y}. \end{equation} (4.2)

    Multiplying the left hand side of (4.2) by \rho(t{\bf v})t^{d-1} and integrating in t\in(0, r) and in {\bf v}\in \Sigma\cap \mathbb{S}^{d-1} , we get

    \begin{equation*} \begin{aligned} \int_{0}^{r}\int_{\Sigma\cap \mathbb{S}^{d-1}} \int_{\Omega\cap B_{\frac{r}{4}}}|{\bf u}( {\bf y})\cdot {\bf v}|^{p}\rho(t{\bf v}) t^{d-1}d {\bf y}\, d\sigma({\bf v}) \, dt = \int_{\Omega\cap B_{\frac{r}{4}}}\int_{\Sigma\cap B_{r}}|{\bf u}( {\bf y})\cdot \frac{{\bf z}}{| {\bf z}|}|^{p}\rho( {\bf z})d {\bf z} \, d {\bf y} \, . \end{aligned} \end{equation*}

    Using Lemma 3.4, we observe that

    \begin{equation} \begin{aligned} & \int_{\Omega\cap B_{\frac{r}{4}}}\int_{\Sigma\cap B_{r}}|{\bf u}( {\bf y})\cdot \frac{{\bf z}}{| {\bf z}|}|^{p}\rho( {\bf z})d {\bf z} \, d {\bf y}\\ &\qquad = \int_{\Omega\cap B_{\frac{r}{4}}}|{\bf u}( {\bf y}) |^{p}\int_{\Sigma\cap B_{r}}\left|\frac{{\bf u}( {\bf y})}{|{\bf u}( {\bf y})|}\cdot \frac{{\bf z}}{| {\bf z}|}\right|^{p}\rho( {\bf z})d {\bf z} \, d {\bf y}\\ &\qquad \geq \left(\int_{0}^{r}t^{d-1}\rho(t) dt\right) \int_{\Omega\cap B_{{r \over 4}}}|{\bf u}( {\bf y}) |^{p} \int_{\Sigma\cap \mathcal{S}^{d-1}}\left|\frac{{\bf u}( {\bf y})}{|{\bf u}( {\bf y})|}\cdot {\bf w}\right|^{p}d \mathcal{H}^{d-1}({\bf w}) \, d {\bf y} \\ &\qquad \geq c_{0} \left(\int_{B_{r}}\rho( {{ \boldsymbol \xi}})d {{ \boldsymbol \xi}}\right) \int_{\Omega\cap B_{\frac{r}{4}}}|{\bf u}( {\bf y})|^{p}d {\bf y}. \end{aligned} \end{equation} (4.3)

    Similarly, we have

    \begin{equation} \begin{aligned} \int_{0}^{r}&\int_{\Sigma\cap \mathbb{S}^{d-1}} \int_{\Omega\cap B_{3r}} \frac{|({\bf u}( {\bf y} +t{\bf v} )-{\bf u}( {\bf y}))\cdot {\bf v}|^{p}}{t^{p}}\rho(t{\bf v}) t^{d-1}d {\bf y}\, d\sigma({\bf v})\, d t\\ & = \int_{\Omega\cap B_{3r}}\int_{\Sigma\cap B_{r}}\frac{|({\bf u}( {\bf y} +{\bf z} )-{\bf u}( {\bf y}))\cdot \frac{{\bf z}}{| {\bf z}|}|^{p}}{| {\bf z}|^{p}}\rho(|{\bf z}|) d{\bf z} d {\bf y}\, \\ &\leq \int_{\Omega\cap B_{4r}}\int_{\Omega\cap B_{4r}}| {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}. \end{aligned} \end{equation} (4.4)

    Combining inequalities (4.2)–(4.4) we obtain that

    \begin{equation} c_{0} \int_{\Omega\cap B_{\frac{r}{4}}}|{\bf u}( {\bf y})|^{p}d {\bf y} \leq \, \frac{r^{p}}{ \int_{B_r} \rho( {{ \boldsymbol \xi}})d {{ \boldsymbol \xi}} }\, \int_{\Omega\cap B_{4r}}\int_{\Omega\cap B_{4r}}| {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x} \end{equation} (4.5)

    for some positive constant c_{0} which only depends on d, p and the Lipschitz constant of the domain. In particular, the estimate (4.5) holds true at all boundary points {\overline{ \boldsymbol \eta}} \in \partial \Omega .

    The next argument is used in the proof of [28, Lemma 5.1]. By applying standard covering argument, it follows from the inequality (4.5) that there exist positive constants \epsilon_{0}\in (0, 1/(2\sqrt{10})) and C with the property that for all r\in(0, r_{0}), such that for all {\bf u}\in L^{p}(\Omega; \mathbb{R}^{d}) that vanishes in \Omega_{{r}/{\sqrt{10}}}

    \begin{equation} \int_{\Omega\setminus\Omega_{2\epsilon_{0}r }} |{\bf u}|^{p}d {\bf x}\leq C \, \frac{r^{p}}{ \int_{B_r} \rho( {{ \boldsymbol \xi}})d {{ \boldsymbol \xi}} }\int_{\Omega}\int_{\Omega} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}. \end{equation} (4.6)

    The positive constants \epsilon_{0} and C depend only on p and the Lipschitz character of the boundary of \Omega. For ease of calculation, set \tilde{r} = 2r/\sqrt{10} . Then \tilde{r}/2 = r/\sqrt{10} .

    Now let {\bf u}\in L^{p}(\Omega; \mathbb{R}^{d}), and let \phi\in C^{\infty}(\Omega) be such that: \phi({\bf x}) = 0 , if {\bf x}\in \Omega_{\tilde{r}/2} ; 0\leq \phi({\bf x})\leq 1, if {\bf x}\in \Omega_{\tilde{r}/4}\setminus \Omega_{\tilde{r}/2} ; \phi({\bf x}) = 1, if {\bf x}\in \Omega\setminus \Omega_{\tilde{r}/4} and |\nabla \phi|\leq C/{r} on \Omega. Applying (4.6) to the vector field \phi({\bf x}){\bf u}({\bf x}) , we obtain that

    \begin{equation*} \begin{aligned} \int_{\Omega\setminus \Omega_{\epsilon_{0} \, r }} |{\bf u}|^{p}d {\bf x} &\leq C\frac{r^{p}}{ \int_{B_r} \rho( {{ \boldsymbol \xi}})d {{ \boldsymbol \xi}} }\int_{\Omega}\int_{\Omega} | {\mathscr D}(\phi{\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}. \end{aligned} \end{equation*}

    We may rewrite {\mathscr D}(\phi{\bf u}) as follows

    {\mathscr D}(\phi{\bf u})( {\bf x}, {\bf y}) = (\phi( {\bf x}) + \phi( {\bf y})) {\mathscr D}({\bf u})( {\bf x}, {\bf y}) - \left({\phi( {\bf x}){\bf u}( {\bf y}) - \phi( {\bf y}){\bf u}( {\bf x})\over | {\bf y}- {\bf x}|}\right)\cdot {( {\bf y}- {\bf x})\over | {\bf y}- {\bf x}|}.

    It then follows that

    \begin{aligned} & \int_{\Omega}\int_{\Omega} | {\mathscr D}(\phi{\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x} \\ &\quad \leq 2^{p-1} \int_{\Omega}\int_{\Omega} |[\phi( {\bf x}) + \phi( {\bf y})] {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x} \\ &\qquad + 2^{p-1}\int_{\Omega}\int_{\Omega} \left|{\phi( {\bf x}){\bf u}( {\bf y}) - \phi( {\bf y}){\bf u}( {\bf x})\over | {\bf y}- {\bf x}|}\cdot {( {\bf y}- {\bf x})\over | {\bf y}- {\bf x}|}\right|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ &\quad = 2^{p-1} \left(I_{1} + I_{2}\right). \end{aligned}

    The first term I_{1} can be easily estimated as

    \begin{aligned} I_{1} & = \int_{\Omega}\int_{\Omega} |[\phi( {\bf x}) + \phi( {\bf y})] {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ & \leq 2 \int_{\Omega}\int_{\Omega} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}. \end{aligned}

    Let us estimate the second term, I_{2} . We first break it into three integrals.

    \begin{aligned} I_{2} & = \int_{\Omega}\int_{\Omega} \left|{\phi( {\bf x}){\bf u}( {\bf y}) - \phi( {\bf y}){\bf u}( {\bf x})\over | {\bf y}- {\bf x}|}\cdot {( {\bf y}- {\bf x})\over | {\bf y}- {\bf x}|}\right|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ & = \iint_{A} + \iint_{B} + \iint_{C} \end{aligned}

    where A = \Omega\setminus \Omega_{\tilde{r}/4} \times \Omega\setminus \Omega_{\tilde{r}/4} , B = (\Omega\setminus \Omega_{\tilde{r}/8})\times \Omega_{\tilde{r}/4} \cup \left(\Omega_{\tilde{r}/4} \times (\Omega\setminus \Omega_{\tilde{r}/8})\right) and C = \Omega\times\Omega \setminus (A\cup B) . We estimate each of these integrals. Let us begin with the simple one: \iint_{A} . After observing that \phi({\bf x}) = \phi({\bf y}) = 1 for all {\bf x}, {\bf y}\in \Omega\setminus \Omega_{\tilde{r}/4}, we have that

    \iint_{A} = \int_{\Omega\setminus \Omega_{\tilde{r}/4}}\int_{\Omega\setminus \Omega_{\tilde{r}/4}} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x},

    and the latter is bounded by the semi norm. Next, we note that set B is symmetric with respect to the diagonal, and as a result,

    \iint_{B} = 2 \int_{\Omega\setminus \Omega_{\tilde{r}/8}} \int_{\Omega_{\tilde{r}/4}}

    and when ({\bf x}, {\bf y}) \in (\Omega\setminus \Omega_{\tilde{r}/8})\times \Omega_{\tilde{r}/4} , we have \phi({\bf x}) = 1 , and so we have

    \begin{aligned} \iint_{B} & = 2\int_{\Omega\setminus \Omega_{\tilde{r}/8}} \int_{\Omega_{\tilde{r}/4}} \left|{{\bf u}( {\bf y}) - \phi( {\bf y}){\bf u}( {\bf x})\over | {\bf y}- {\bf x}|}\cdot {( {\bf y}- {\bf x})\over | {\bf y}- {\bf x}|}\right|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ &\leq 2^{p}\int_{\Omega\setminus \Omega_{\tilde{r}/8}} \int_{\Omega_{\tilde{r}/4}} \left|\phi( {\bf y}) {\mathscr D}({\bf u})( {\bf x}, {\bf y})\right|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ &\quad + 2^{p}\int_{\Omega\setminus \Omega_{\tilde{r}/8}} \int_{\Omega_{\tilde{r}/4}} \left|{{\bf u}( {\bf y})\over | {\bf y}- {\bf x}|}\right|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ &\leq 2^{p}\int_{\Omega\setminus \Omega_{\tilde{r}/8}} \int_{\Omega_{\tilde{r}/4}}\left| {\mathscr D}({\bf u})( {\bf x}, {\bf y})\right|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ &\quad + \frac{2^{4p}}{r^{p}}\int_{\Omega\setminus \Omega_{\tilde{r}/8}} \int_{\Omega_{\tilde{r}/4}} |{\bf u}( {\bf y})|^p\rho( {\bf x}- {\bf y})d {\bf y}\, d{\bf x}\\ \end{aligned}

    where we have used the fact that \text{dist}(\Omega\setminus \Omega_{\tilde{r}/8}, \Omega_{\tilde{r}/4}) = \tilde{r}/{8} . As a consequence we have that

    \iint_{B} \leq 2^{p} \int_{\Omega}\int_{\Omega} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p \rho( {\bf y}- {\bf x})d {\bf y}\, d {\bf x} + \frac{2^{4p}}{r^{p}} \left(\int_{|{\bf h}| > \tilde{r}/{8}} \rho({\bf h}) d{\bf h} \right) \int_{\Omega_{\tilde{r}/4}} |{\bf u}( {\bf y})|^{p} d {\bf y}.

    To estimate the integral on C , we first observe that for any ({\bf x}, {\bf y}) \in C , then \text{dist}({\bf x}, \partial \Omega)\geq \tilde{r}/{8} and \text{dist}({\bf y}, \partial \Omega)\geq \tilde{r}/{8} . Using this information, adding and subtracting \phi({\bf x}){\bf u}({\bf x}) we can then estimate as follows:

    \begin{aligned} \iint_{C} &\leq 2^{p-1} \iint_{C} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p \rho( {\bf y}- {\bf x})d {\bf y}\, d {\bf x} \\ &\quad + 2^{p-1}\iint_{C}|{\bf u}( {\bf x})|^{p} \frac{|\phi( {\bf x} )-\phi( {\bf y})|^{p}}{| {\bf x}- {\bf y}|^{p}}\rho(| {\bf x}- {\bf y}|)d {\bf y}\, d{\bf x} \\ & \leq 2^{p-1} \int_{\Omega}\int_{\Omega} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p \rho( {\bf y}- {\bf x})d {\bf y}\, d {\bf x}\\ & \quad + \frac{C}{r^{p}} \int_{B_{R}} \rho({\bf h}) d{\bf h} \int_{\Omega_{{r\over8}}}|{\bf u}( {\bf x})|^{p} d {\bf x} \end{aligned}

    where we used the estimate |\nabla \phi| \leq {C\over r} , and denoted R = \text{diam}(\Omega) .

    We then conclude that there exists a universal constant C > 0 such that for any r small

    \begin{aligned} \int_{\Omega\setminus \Omega_{\epsilon_{0} \, r }} |{\bf u}|^{p}d {\bf x} & \leq C \left({r^{p}\over \int_{B_{r}}\rho(|{ {\bf y}}|)d {\bf y}} \int_{\Omega}\int_{\Omega} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p \rho( {\bf y}- {\bf x})d {\bf y}\, d {\bf x} \right.\\ &\qquad +\left. \frac{1}{r^{p}}\int_{B_{R}} \rho({\bf h}) d{\bf h} \int_{\Omega_{\tilde{r}/{8}}} |{\bf u}|^{p} d {\bf x}\right) . \end{aligned}

    It then follows that

    \begin{equation*} \begin{aligned} \int_{\Omega}|{\bf u}|^{p}d {\bf x} & = \int_{\Omega_{\epsilon_{0}r}}|{\bf u}|^{p}d {\bf x} + \int_{\Omega\setminus \Omega_{\epsilon_{0}r}}|{\bf u}|^{p}d {\bf x} \\ &\leq \int_{\Omega_{\epsilon_{0} r}}|{\bf u}|^{p}d {\bf x} + C\, {r^{p}\over \int_{B_{r}}\rho(|{ {\bf y}}|)d {\bf y}} \int_{\Omega}\int_{\Omega} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p \rho( {\bf y}- {\bf x})d {\bf y}\, d {\bf x} \\ &\quad+ C {\|\rho\|_{L^{1}(B_{R})}\over r^{p}}\int_{\Omega_{\tilde{r}/8}}|{\bf u}|^{p}d {\bf x}. \end{aligned} \end{equation*}

    We hence complete the proof of Lemma 4.2 after choosing \epsilon_{0} sufficiently small, say for example \epsilon_{0} < {1}/{4\sqrt{10}} , that

    \begin{equation*} \int_{\Omega}|{\bf u}|^{p}d {\bf x} \leq C(r)\int_{\Omega_{\epsilon_{0}r}}|{\bf u}|^{p}d {\bf x} + C\, {r^{p}\over \int_{B_{r}(0)}\rho(|{ {\bf y}}|)d {\bf y}} \int_{\Omega}\int_{\Omega} | {\mathscr D}({\bf u})( {\bf x}, {\bf y})|^p \rho( {\bf y}- {\bf x})d {\bf y}\, d {\bf x}, \end{equation*}

    as desired.

    Let {\bf u}_{n} be a bounded sequence in \mathcal{S}_{\rho, p} (\Omega) . Let \phi_{j} \in C^{\infty}_{0}(\Omega) such that \phi_{j} \equiv 1 in \Omega_{1/j} . Then the sequence \{\phi_{j}{\bf u}_{n}\}_{n} is bounded in \mathcal{S}_{\rho, p} (\mathbb{R}^{d}) , and so by Theorem 3.1, \phi_{j}{\bf u}_{n} is precompact in \Omega . In particular, \{{\bf u}_{n}\} is relatively compact in L^{p}(\Omega_{j}) . From this one can extract a subsequence {\bf u}_{n_{j}} such that {\bf u}_{n_{j}} \to {\bf u} in L^{p}_{loc}(\Omega) . It is easy to see that {\bf u}\in L^{p}(\Omega) . In fact, using the pointwise convergence and Fatou's lemma, we can see that {\bf u}\in \mathcal{S}_{\rho, p}(\Omega) . What remains is to show that {{\bf u}_{n_{j}}} \to {\bf u} in L^{p}(\Omega) . To that end, we apply Lemma 4.2 for the function {\bf u}_{n_{j}} - {\bf u} , to obtain that

    \int_{\Omega}|{\bf u}_{n_{j}} - {\bf u}|^{p}d {\bf x} \leq C_{1}(r) \int_{\Omega_{\epsilon_{0}r}}|{\bf u}_{n_{j}} - {\bf u}|d {\bf x} + C_{2} \frac{r^{p}}{ \int_{B_{r}}\rho({\bf h}) d{\bf h} } |{\bf u}_{n_{j}} - {\bf u}|^{p}_{\mathcal{S}_{\rho, p}(\Omega)}

    for all small r . We now fix r and let j\to \infty to obtain that

    \limsup\limits_{j\to \infty}\int_{\Omega}|{\bf u}_{n_{j}} - {\bf u}|^{p}d {\bf x} \leq C \frac{r^{p}}{ \int_{B_{r}}\rho({\bf h}) d{\bf h} } (1 + |{\bf u}|^{p}_{\mathcal{S}_{\rho, p}}).

    We then let r\to 0 , to obtain that \limsup_{j\to \infty}\int_{\Omega}|{\bf u}_{n_{j}} - {\bf u}|^{p}d {\bf x} = 0.

    Arguing as above and by Proposition 3.5, we have that there is a subsequence {\bf u}_{n_{j}} \to u in L^{p}_{loc}(\Omega) , and that u\in \mathcal{S}_{\rho, p}(\Omega) . To conclude, we apply Lemma 4.2 for the function {\bf u}_{n_{j}} - {\bf u} corresponding to \rho_{n_{j}} to obtain

    \int_{\Omega}|{\bf u}_{n_{j}} - {\bf u}|^{p}d {\bf x} \leq C_{1}(r) \int_{\Omega_{\epsilon_{0}r}}|{\bf u}_{n_{j}} - {\bf u}|d {\bf x} + C_{2} \frac{r^{p}}{ \int_{B_{r}}\rho_{n_{j}}({\bf h}) d{\bf h} } |{\bf u}_{n_{j}} - {\bf u}|^{p}_{\mathcal{S}_{\rho_{n_j}, p}(\Omega)} .

    By assumption \rho_{n_j}\leq C \rho and so |{\bf u}_{n_{j}} - {\bf u}|^{p}_{\mathcal{S}_{\rho_{n_j}, p}(\Omega)}\leq C |{\bf u}_{n_{j}} - {\bf u}|^{p}_{\mathcal{S}_{\rho, p}(\Omega)} . We then let j\to \infty and apply the weak convergence of \rho_{n} to obtain that

    \limsup\limits_{j\to \infty}\int_{\Omega}|{\bf u}_{n_{j}} - {\bf u}|^{p}d {\bf x} \leq C \frac{r^{p}}{ \int_{B_{r}}\rho({\bf h}) d{\bf h} } (1 + |{\bf u}|^{p}_{\mathcal{S}_{\rho, p}}).

    Finally, we let r\to 0 to conclude the proof.

    We recall that given V\subset L^{p}(\Omega; \mathbb{R}^{d}) satisfying the hypothesis of the corollary, there exists a constant P_{0} such that for any {\bf u}\in V,

    \begin{equation} \int_{\Omega} |{\bf u}|^{p}d {\bf x} \leq P_{0} \int_{\Omega}\int_{\Omega} \rho( {\bf y} - {\bf x})\left|\frac{({\bf u}( {\bf y}) - {\bf u}({ {\bf x}}))}{| {\bf y}- {\bf x}|} \cdot \frac{( {\bf y} - {\bf x})}{| {\bf y} - {\bf x}|}\right|^{p}d {\bf y} d {\bf x}. \end{equation} (4.7)

    This result is proved in [12] or [23]. We take P_{0} to be the best constant. We claim that given any \epsilon > 0 , there exists N = N(\epsilon) \in \mathbb{N} such that for all n\geq N , (1.15) holds for C = P_{0} + \epsilon . We prove this by contradiction. Assume otherwise and that there exists C > P_{0} such that for every n , there exists {\bf u}_{n}\in V\cap L^{p}(\Omega; \mathbb{R}^{d}) , \|{\bf u}_{n}\|_{L^{p}} = 1 , and

    \int_{\Omega}\int_{\Omega} \rho_{n}( {\bf y} - {\bf x})\left|\frac{({\bf u}_{n}( {\bf y}) - {\bf u}_{n}({ {\bf x}}))}{| {\bf y}- {\bf x}|} \cdot \frac{( {\bf y} - {\bf x})}{| {\bf y} - {\bf x}|}\right|^{p}d {\bf y} d {\bf x} < \frac{1}{C}.

    By Theorem 1.4, {\bf u}_{n} is precompact in L^{p}(\Omega; \mathbb{R}^{d}) and therefore any limit point {\bf u} will have \|{\bf u}\|_{L^{p}} = 1 , and will be in V\cap L^{p}(\Omega; \mathbb{R}^{d}) . Moreover, following the same procedure as in the proof of Proposition 3.5, we obtain that

    \begin{aligned} \int_{\Omega}\int_{\Omega} &\rho( {\bf y} - {\bf x})\left|\frac{({\bf u}( {\bf y}) - {\bf u}({ {\bf x}}))}{| {\bf y}- {\bf x}|} \cdot \frac{( {\bf y} - {\bf x})}{| {\bf y} - {\bf x}|}\right|^{p}d {\bf y} d {\bf x} \\ &\leq \liminf\limits_{n\to \infty}\int_{\Omega}\int_{\Omega} \rho_{n}( {\bf y} - {\bf x})\left|\frac{({\bf u}_{n}( {\bf y}) - {\bf u}_{n}({ {\bf x}}))}{| {\bf y}- {\bf x}|} \cdot \frac{( {\bf y} - {\bf x})}{| {\bf y} - {\bf x}|}\right|^{p}d {\bf y} d {\bf x} \leq {1 \over C } < {1\over P_{0}} \end{aligned}

    which gives the desired contradiction since P_{0} is the best constant in (4.7).

    In this work, we have presented a set of sufficient conditions that guarantee a compact inclusion of a set of L^{p} -vector fields in the Banach space of L^{p} vector fields. The criteria are nonlocal and given with respect to nonlocal interaction kernels that may not be necessarily radially symmetric. We note that, in addition to the mathematical generality, relaxing the radial symmetry assumption on nonlocal interactions can be useful when modeling anisotropic behavior and directional transport. The L^{p} -compactness is established for a sequence of vector fields where the nonlocal interactions involve only part of their components, so that the results and discussions represent a significant departure from those known for scalar fields. It is not clear yet to what extent the conditions assumed here can be weakened to reach the same conclusions. In this regard, there are still some outstanding questions in relation to the set of minimal conditions on the interaction kernel as well as on the set of vector fields that imply L^{p} -compactness. An application of the compactness result that will be explored elsewhere includes designing approximation schemes for nonlocal system of equations of peridynamic-type similar to the one done in [34] for nonlocal equations.

    The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.

    Q. Du's research is partially supported by US National Science Foundation grant DMS-1937254 and DMS-2309245.

    T. Mengesha's research is supported by US National Science Foundation grant DMS-1910180 and DMS-2206252.

    X. Tian's research is supported by US National Science Foundation grant DMS-2111608 and DMS-2240180.

    The authors thank Zhaolong Han for helping improve the proof of Lemma 4.2.

    The authors declare no conflicts of interest.

    Compactness in L^{p}_{loc} topology

    The following theorem as well as the proof we present here is inspired by the compactness result proved in [18] (see also [17]) for scalar functions which uses the more flexible nonintegrability condition of the {\rho({\bf z}) \over |{\bf z}|^{p}} than the one stated in (1.11).

    Theorem A.1 ( L^{p}_{loc} compactness). Suppose that 1\leq p < \infty . Let \rho \in L^{1}(\mathbb{R}^{d}) be a nonnegative radial function satisfying

    \begin{equation} \lim\limits_{\delta\to 0 } \int_{|{\bf z}| > \delta} {\rho({\bf z})\over |{\bf z}|^{p}}d{\bf z} = \infty. \end{equation} (A.1)

    Suppose also that \{{\bf u}_{n}\} is a sequence of vector fields that is bounded in \mathcal{S}_{\rho, p}(\mathbb{R}^{d}) . Then for any D\subset \mathbb{R}^{d} open and bounded, the sequence \{{\bf u}_{n}|_{D}\} is precompact in L^{p}(D; \mathbb{R}^{d}) .

    As stated earlier, for radial functions with compact support, condition (9.1) is weaker than (1.11). Indeed, (1.11) implies that \rho({\bf z}) | {\bf z}|^{-p} is not integrable near { \bf 0} which implies (9.1). One the other hand, the kernel \rho({\bf z}) = |{\bf z}|^{-d-p} \chi_{B_{1}({ \bf 0})}({\bf z}) satisfies (A.1) but not (1.11).

    Similar to the argument we gave in Section 2, the proof of the theorem will make use of the following variant of the Riesz-Fréchet-Kolomogorov theorem [6,20].

    Lemma A.2. ([20, Lemma 5.4]) Let the sequence \{\mathbb{G}^\delta\}_{\delta > 0}\subset L^1(\mathbb{R}^{d}; \mathbb{R}^{d\times d}) be an approximation to the identity. That is

    \forall \delta > 0, \int_{\mathbb{R}^{d}} \mathbb{G}^\delta ( {\bf x})d {\bf x} = \mathbb{I}_d, \quad \mathit{\text{for any $r > 0$, }} \lim\limits_{\delta \to 0}\int_{| {\bf x}| > r} \mathbb{G}^\delta ( {\bf x})d {\bf x} = {\bf 0}.

    If \{{\bf f}_n\}_{n} is a bounded sequence in L^{p}(\mathbb{R}^{d}; \mathbb{R}^{d}) and

    \lim\limits_{\delta \to 0} \limsup\limits_{n\to \infty} \|{\bf f}_{n} - \mathbb{G}^\delta \ast {\bf f}_{n}\|_{L^{p}} = 0,

    then for any open and bounded subset D of \mathbb{R}^{d} the sequence \{{\bf f}_{n}\} is relatively compact in L^{p}(D; \mathbb{R}^{d}) .

    Proof of Theorem A.1. From the assumption we have

    \begin{equation} \sup\limits_{n\geq 1} \|{\bf u}_{n}\|_{L^{p}}^{p} + \sup\limits_{n\geq 1}\int_{\mathbb{R}^{d}}\int_{\mathbb{R}^{d}}\rho( {\bf x}'- {\bf x})\left| {\mathscr D}({\bf u}_n)( {\bf x}, {\bf x}')\right|^{p}d {\bf x}'d {\bf x} < \infty. \end{equation} (A.2)

    Let \Gamma^{\delta}({\bf z}) = {\rho({\bf z}) \over |{\bf z}|^{p}} \chi_{\complement B_{\delta}}({\bf z}) . Then for each \delta , \Gamma^\delta \in L^1(\mathbb{R}^{d}) and is radial, since \rho is radial. Moreover, by assumption on \rho (A.1), \|\Gamma^\delta\|_{L^1} \to \infty as \delta \to 0. We next introduce the following sequence of integrable matrix functions

    \mathbb{G}^{\delta}({\bf z}) = {d \, \Gamma^{\delta}({\bf z})\over \|\Gamma^\delta\|_{L^1} } {{\bf z}\otimes {\bf z} \over |{\bf z}|^{2}}.

    Notice that since \Gamma^\delta is radial, we have

    \int_{\mathbb{R}^{\delta}} {\Gamma^{\delta}({\bf z}) z_i^{2} \over |{\bf z}|^{2}}d{\bf z} = { \|\Gamma^\delta\|_{L^1}\over d}, \quad i = 1, \cdots, d.

    As a consequence \{\mathbb{G}^{\delta}\} is an approximation to the identity. Now for each n we have

    \begin{aligned} \| {\bf u}_{n} - \mathbb{G}^{\delta} \ast{\bf u}_n\|_{L^{p}}^{p} & = \int_{\mathbb{R}^{d}}\left|\int_{\mathbb{R}^{d}}\mathbb{G}^{d} ( {\bf y}- {\bf x})({\bf u}_n( {\bf x}) - {\bf u}_n( {\bf y})) d {\bf y} \right|^{p} d {\bf x}\\ & \leq d^{p} \int_{\mathbb{R}^{d}}\left|\int_{\mathbb{R}^{d}} {\Gamma^{\delta}({\bf z})\over \|\Gamma^\delta\|_{L^1} }\left|{ {\bf z}\over | {\bf z}|}\cdot({\bf u}_n( {\bf x}) - {\bf u}_n( {\bf z} + {\bf x})) \right|d {\bf y} \right|^{p} d {\bf x}\\ &\leq d^{p} \int_{\mathbb{R}^{d}}\int_{\mathbb{R}^{d}} \left|{ {\bf z}\over | {\bf z}|}\cdot({\bf u}_n( {\bf x}) - {\bf u}_n( {\bf z} + {\bf x})) \right|^{p} {\Gamma^{\delta}({\bf z})\over \|\Gamma^\delta\|_{L^1} } d {\bf z} d {\bf x}\\ &\leq {d^{p}\over \|\Gamma^\delta\|_{L^1} } \int_{\mathbb{R}^{d}}\int_{\mathbb{R}^{d}} \left|{ {\bf z}\over | {\bf z}|}\cdot({\bf u}_n( {\bf x}) - {\bf u}_n( {\bf z} + {\bf x})) \right|^{p} {\rho({\bf z}) \over |{\bf z}|^{p}} d {\bf z} d {\bf x}\\ & = {d^{p}\over \|\Gamma^\delta\|_{L^1} }|{\bf u}_{n}|^p_{\mathcal{S}_{\rho, p}(\mathbb{R}^{d})}. \end{aligned}

    By assumption on the sequence \{{\bf u}_n\} (A.2), we have that for all n ,

    \| {\bf u}_{n} - \mathbb{G}^{\delta} \ast{\bf u}_n\|_{L^{p}}^{p} \leq C \, {d^{p}\over \|\Gamma^\delta\|_{L^1}}.

    We take the limit as \delta\to 0 (uniformly in n ) and use Lemma A.2 to conclude that {\bf u}_n is compact in L^{p}(\Omega; \mathbb{R}^{d}).



    [1] S. Zhang, C. Wu, L. Liu, et al., Optical coherence tomography angiography of the peripapillary retina in primary angle-closure glaucoma, Am. J. Ophthalmol., 182 (2017), 194.
    [2] J. Sara, A. Rouzbeh and V. H. Barocas, Contribution of different anatomical and physiologic factors to iris contour and anterior chamber angle changes during pupil dilation: theoretical analysis, Invest. Ophth. Vis. Sci., 54 (2013), 2977–2984.
    [3] F. Zhang and H. Chen, Numerical investigation of laser iridotomy influence on shear stress exerted on corneal endothelial cells, J. Med. Biomech., (2016).
    [4] L. W. Schwartz and G. L. Spaeth, Argon laser iridotomy in primary angle-closure or pupillary block glaucoma, Laser. Surg. Med., 1 (2010), 153–164.
    [5] N. Satoru and M. Akira, Three-year outcome of Descemet stripping automated endothelial keratoplasty for bullous keratopathy after argon laser iridotomy, Cornea, 33 (2014), 780–784.
    [6] O. V. Danilenko and A. V. Bol'Shunov, Laser iridectomy and anatomical and functional parameters in primary angle-closure glaucoma, Vestnik Oftalmologii, 130 (2014), 54.
    [7] O. Abouali, A. Modareszadeh, A. Ghaffarieh, et al., Investigation of saccadic eye movement effects on the fluid dynamic in the anterior chamber, J. Biomech. Eng., 134 (2012), 021002.
    [8] E. H. Ooi and E. Y. K. Ng, Effects of natural convection within the anterior chamber on the ocular heat transfer, Int. J. Numer. Meth. Bio. Eng., 27 (2015), 408–423.
    [9] Y. Yasuaki, U. Toshihiko, J. Takeshi, et al., Effect of anterior chamber depth on shear stress exerted on corneal endothelial cells by altered aqueous flow after laser iridotomy, Invest. Ophth. Vis. Sci., 51 (2010), 1956.
    [10] R. Avtar and R. Srivastava, Aqueous outflow in Schlemm's canal, Appl. Math. Comput., 174 (2006), 316–328.
    [11] T. R. Crowder and V. J. Ervin, Numerical simulations of fluid pressure in the human eye, Appl. Math. Comput., 219 (2013), 11119–11133.
    [12] E. C. Huang and V. H. Barocas, Accommodative microfluctuations and iris contour, J. Vision, 6 (2006), 653.
    [13] W. Wang, X. Qian, H. Song, et al., Fluid and structure coupling analysis of the interaction between aqueous humor and iris, Biomed. Eng. Online, 15 (2016), 133.
    [14] Y. Yasuaki, U. Toshihiko, S. Katsumi, et al., Demonstration of aqueous streaming through a laser iridotomy window against the corneal endothelium, Arch. Ophthalmol., 124 (2006), 387.
    [15] J. J. Heys, V. H. Barocas and M. J. Taravella, Modeling passive mechanical interaction between aqueous humor and iris, J. Biomech. Eng., 123 (2001), 540.
    [16] H. Y. Jie, J. H. Heo, H. M. Kim, et al., Effects of Argon Laser Iridotomy on the Corneal Endothelium of Pigmented Rabbit Eyes, Korean J. Ophthalmol., 28 (2014), 76–82.
    [17] P. X. Wang, V. T. C. Koh and L. S. Chee, Laser iridotomy and the corneal endothelium: a systemic review, Acta Ophthalmol., 92 (2015), 604–616.
    [18] R. Y. Lee, K. Toshimitsu, N. Q. Cui, et al., Association between baseline angle width and induced angle opening following prophylactic laser peripheral iridotomy, Invest. Ophth. Vis. Sci., 54 (2013), 3763–3770.
  • This article has been cited by:

    1. Niloofar Hassannejad, Abbas Bahador, Nasim Hayati Rudbari, Mohammad Hossein Modarressi, Kazem Parivar, In vivo antibacterial activity of Zataria multiflora Boiss extract and its components, carvacrol, and thymol, against colistin‐resistant Acinetobacter baumannii in a pneumonic BALB/c mouse model , 2019, 120, 0730-2312, 18640, 10.1002/jcb.28908
    2. Umar Ndagi, Abubakar A. Falaki, Maryam Abdullahi, Monsurat M. Lawal, Mahmoud E. Soliman, Antibiotic resistance: bioinformatics-based understanding as a functional strategy for drug design, 2020, 10, 2046-2069, 18451, 10.1039/D0RA01484B
    3. Saima Muzammil, Mohsin Khurshid, Iqra Nawaz, Muhammad Hussnain Siddique, Muhammad Zubair, Muhammad Atif Nisar, Muhammad Imran, Sumreen Hayat, Aluminium oxide nanoparticles inhibit EPS production, adhesion and biofilm formation by multidrug resistant Acinetobacter baumannii, 2020, 36, 0892-7014, 492, 10.1080/08927014.2020.1776856
    4. Kaoru YAMABE, Yukio ARAKAWA, Masaki SHOJI, Mitsuko ONDA, Katsushiro MIYAMOTO, Takahiro TSUCHIYA, Yukihiro AKEDA, Kuniko TERADA, Kazunori TOMONO, Direct anti-biofilm effects of macrolides on Acinetobacter baumannii: comprehensive and comparative demonstration by a simple assay using microtiter plate combined with peg-lid, 2020, 41, 0388-6107, 259, 10.2220/biomedres.41.259
    5. Kangjian Zhang, Xiaoyan Yang, Jiali Yang, Xia Qiao, Feng Li, Xiaoming Liu, Jun Wei, Lixin Wang, Alcohol dehydrogenase modulates quorum sensing in biofilm formations of Acinetobacter baumannii, 2020, 148, 08824010, 104451, 10.1016/j.micpath.2020.104451
    6. Anthony Adegoke, Adekunle Faleye, Gulshan Singh, Thor Stenström, Antibiotic Resistant Superbugs: Assessment of the Interrelationship of Occurrence in Clinical Settings and Environmental Niches, 2016, 22, 1420-3049, 29, 10.3390/molecules22010029
    7. Vijay Kothari, Chinmayi Joshi, Pooja Patel, Deepa Shahi, Charmi Mehta, Bhumika Prajapati, Sweta Patel, Dipeeka Mandaliya, Sriram Seshadri, 2018, 9780128136676, 325, 10.1016/B978-0-12-813667-6.00008-5
    8. K. Saipriya, C.H. Swathi, K.S. Ratnakar, V. Sritharan, Quorum‐sensing system in Acinetobacter baumannii : a potential target for new drug development , 2020, 128, 1364-5072, 15, 10.1111/jam.14330
    9. Yoshinori Sato, Yuka Unno, Tsuneyuki Ubagai, Yasuo Ono, Hendrik W. van Veen, Sub-minimum inhibitory concentrations of colistin and polymyxin B promote Acinetobacter baumannii biofilm formation, 2018, 13, 1932-6203, e0194556, 10.1371/journal.pone.0194556
    10. Fatemeh Hemmati, Roya Salehi, Reza Ghotaslou, Hossein Samadi Kafil, Alka Hasani, Pourya Gholizadeh, Roghayeh Nouri, Mohammad Ahangarzadeh Rezaee,

    Quorum Quenching: A Potential Target for Antipseudomonal Therapy

    , 2020, Volume 13, 1178-6973, 2989, 10.2147/IDR.S263196
    11. Niharika G. Jha, Daphika S. Dkhar, Sumit K. Singh, Shweta J. Malode, Nagaraj P. Shetti, Pranjal Chandra, Engineered Biosensors for Diagnosing Multidrug Resistance in Microbial and Malignant Cells, 2023, 13, 2079-6374, 235, 10.3390/bios13020235
    12. Evgenia Maslova, Lara Eisaiankhongi, Folke Sjöberg, Ronan R. McCarthy, Burns and biofilms: priority pathogens and in vivo models, 2021, 7, 2055-5008, 10.1038/s41522-021-00243-2
    13. Sowndarya Jothipandiyan, Devarajan Suresh, Sankaran Venkatachalam Sankaran, Subbiah Thamotharan, Kumaravel Shanmugasundaram, Preethi Vincent, Saravanan Sekaran, Shanmugaraj Gowrishankar, Shunmugiah Karutha Pandian, Nithyanand Paramasivam, Heteroleptic pincer palladium(II) complex coated orthopedic implants impede the AbaI/AbaR quorum sensing system and biofilm development by Acinetobacter baumannii, 2022, 38, 0892-7014, 55, 10.1080/08927014.2021.2015336
    14. Muhammad Hussnain Siddique, Sumreen Hayat, Saima Muzammil, Asma Ashraf, Arif Muhammad Khan, Muhammad Umar Ijaz, Mohsin Khurshid, Muhammad Afzal, Ecofriendly phytosynthesized zirconium oxide nanoparticles as antibiofilm and quorum quenching agents against Acinetobacter baumannii, 2022, 48, 0363-9045, 502, 10.1080/03639045.2022.2132260
    15. Sirijan Santajit, Nitat Sookrung, Nitaya Indrawattana, Quorum Sensing in ESKAPE Bugs: A Target for Combating Antimicrobial Resistance and Bacterial Virulence, 2022, 11, 2079-7737, 1466, 10.3390/biology11101466
    16. Veronica Lazar, Alina Maria Holban, Carmen Curutiu, Mariana Carmen Chifiriuc, Modulation of Quorum Sensing and Biofilms in Less Investigated Gram-Negative ESKAPE Pathogens, 2021, 12, 1664-302X, 10.3389/fmicb.2021.676510
    17. Rajat Kumar Jha, Rameez Jabeer Khan, Ekampreet Singh, Ankit Kumar, Monika Jain, Jayaraman Muthukumaran, Amit Kumar Singh, An extensive computational study to identify potential inhibitors of Acyl-homoserine-lactone synthase from Acinetobacter baumannii (strain AYE), 2022, 114, 10933263, 108168, 10.1016/j.jmgm.2022.108168
    18. Rajat Kumar Jha, Ekampreet Singh, Rameez Jabeer Khan, Ankit Kumar, Monika Jain, Jayaraman Muthukumaran, Amit Kumar Singh, Droperidol as a potential inhibitor of acyl-homoserine lactone synthase from A. baumannii: insights from virtual screening, MD simulations and MM/PBSA calculations, 2022, 1381-1991, 10.1007/s11030-022-10533-2
    19. Gustavo Mockaitis, Guillaume Bruant, Eugenio Foresti, Marcelo Zaiat, Serge R. Guiot, Physicochemical pretreatment selects microbial communities to produce alcohols through metabolism of volatile fatty acids, 2022, 2190-6815, 10.1007/s13399-022-02383-7
    20. Sagar Kiran Khadke, Jin-Hyung Lee, Yong-Guy Kim, Vinit Raj, Jintae Lee, Assessment of Antibiofilm Potencies of Nervonic and Oleic Acid against Acinetobacter baumannii Using In Vitro and Computational Approaches, 2021, 9, 2227-9059, 1133, 10.3390/biomedicines9091133
    21. Jwan Khidhr Rahman, Akhter Ahmed Ahmed, Aryan R. Ganjo, Trefa Salih Mohamad, Assessment of toxicity, anti-quorum sensing and anti-biofilm production effects of Hypericum triquetrifolium Turra extract on multi-drug resistant Acinetobacter baumannii, 2023, 35, 10183647, 102714, 10.1016/j.jksus.2023.102714
    22. Lizhi Li, Weikun Guan, Yihao Fan, Qin He, Dongsheng Guo, An Yuan, Qingfeng Xing, Yang Wang, Ziqin Ma, Jian Ni, Jia Chen, Qilong Zhou, Yuhong Zhong, Jiating Li, Haibo Zhang, Zinc/carbon nanomaterials inhibit antibiotic resistance genes by affecting quorum sensing and microbial community in cattle manure production, 2023, 387, 09608524, 129648, 10.1016/j.biortech.2023.129648
    23. Seetha Lakshmi Rajangam, Manoj Kumar Narasimhan, Current treatment strategies for targeting virulence factors and biofilm formation in Acinetobacter baumannii , 2024, 19, 1746-0913, 941, 10.2217/fmb-2023-0263
    24. Christiana E. Aruwa, Theolyn Chellan, Nosipho W. S'thebe, Yamkela Dweba, Saheed Sabiu, ESKAPE pathogens and associated quorum sensing systems: New targets for novel antimicrobials development, 2024, 11, 27726320, 100155, 10.1016/j.hsr.2024.100155
    25. Gabriel Gbenga Babaniyi, Babafemi Raphael Babaniyi, Ulelu Jessica Akor, 2023, 978-1-83916-761-4, 59, 10.1039/BK9781837671380-00059
    26. Andrea Marino, Egle Augello, Stefano Stracquadanio, Carlo Maria Bellanca, Federica Cosentino, Serena Spampinato, Giuseppina Cantarella, Renato Bernardini, Stefania Stefani, Bruno Cacopardo, Giuseppe Nunnari, Unveiling the Secrets of Acinetobacter baumannii: Resistance, Current Treatments, and Future Innovations, 2024, 25, 1422-0067, 6814, 10.3390/ijms25136814
    27. Sérgio G. Mendes, Sofia I. Combo, Thibault Allain, Sara Domingues, Andre G. Buret, Gabriela J. Da Silva, Co-regulation of biofilm formation and antimicrobial resistance in Acinetobacter baumannii: from mechanisms to therapeutic strategies, 2023, 42, 0934-9723, 1405, 10.1007/s10096-023-04677-8
    28. Mohammed Mansour Quradha, Alfred Ngenge Tamfu, Mehmet Emin Duru, Selcuk Kucukaydin, Mudassar Iqbal, Abdulkader Moqbel Farhan Qahtan, Rasool Khan, Ozgur Ceylan, Evaluation of HPLC Profile, Antioxidant, Quorum Sensing, Biofilm, Swarming Motility, and Enzyme Inhibition Activities of Conventional and Green Extracts of Salvia triloba, 2024, 2048-7177, 10.1002/fsn3.4580
    29. Abirami Karthikeyan, Manoj Kumar Thirugnanasambantham, Fazlurrahman Khan, Arun Kumar Mani, Bacteria-Inspired Synthesis of Silver-Doped Zinc Oxide Nanocomposites: A Novel Synergistic Approach in Controlling Biofilm and Quorum-Sensing-Regulated Virulence Factors in Pseudomonas aeruginosa, 2025, 14, 2079-6382, 59, 10.3390/antibiotics14010059
    30. Thejaswi Bhat, Manish Kumar, Krishna Kumar Ballamoole, Vijaya Kumar Deekshit, Pavan Gollapalli, Pangenome-based network analysis of Acinetobacter baumannii reveals the landscape of conserved therapeutic targets, 2025, 1381-1991, 10.1007/s11030-025-11252-0
  • Reader Comments
  • © 2019 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(5261) PDF downloads(457) Cited by(6)

Figures and Tables

Figures(11)  /  Tables(2)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog