R E S E A R C H
Open Access
Stability of weak solutions to obstacle
problem in Clifford analysis
Yueming Lu
*and Gejun Bao
*Correspondence:
Department of Mathematics, Harbin Institute of Technology, Harbin, 150001, P.R. China
Abstract
This paper is concerned with the regularity properties of weak solutions to the obstacle problem for Clifford-valued functions. Our main results are a global reverse Hölder inequality and stability of the weak solutions to the obstacle problem.
Keywords: Clifford-valued function; obstacle problem; higher integrability; stability
1 Introduction
Lete,e, . . . ,en be the standard basis ofRnwith the relationeiej+ejei= –δij. Fork=
, , . . . ,n, we denote by Ckn=Ckn(Rn) the linear space of all k-multivectors, spanned
by the reduced products eI = eiei· · ·eik, corresponding to all ordered k-tuples I = (i,i, . . . ,ik), ≤i <i<· · ·<ik≤n. Thus, Clifford algebraCn=kCkn is a graded
algebra, especiallyCn=RandCn=Rn.R⊂C⊂H⊂Cn⊂ · · ·is an increasing chain, whereHis the Hamilton’s algebra of quaternions. Foru∈Cn,ucan be written as
u=
I uIeI=
≤i<···<ik≤n
ui,...,ikei· · ·eik,
where ≤k≤n.
The norm of u ∈Cn is given by |u| = (
IuI)/. Clifford conjugation eα· · ·eαk = (–)keαk· · ·eα. Foru=
IuIeI∈Cn,v=
JvJeJ∈Cn,
u,v=
I
uIeI,
J vJeJ
=
I uIvI
defines the corresponding inner product onCn. Denote by Sc(u) the scalar part ofu, the
coefficient of the elemente, and we also haveu,v= Sc(uv). The Dirac operator used in this paper is given by
D=
n
j=i ej
∂
∂xj. ()
Throughout this paper, we writeD(,Cn) for the space of Clifford-valued functions
whose coefficients are Schwartz distributions on. Forp> , we writeLp(,Ckn) for the space of Clifford-valued functionsu, whose coefficients belong to the usualLp() space.
It is a reflexive Banach space endowed with the norm
up= ˆ
|u|pdx
p .
Also,W,p(,Ck
n) is the Sobolev space of Clifford-valued functionsu, whose coefficients
belong toW,p(). It is a reflexive Banach space endowed with the norm
u,p=up+up,
whereu= (∂∂xu
, . . . ,
∂u
∂xn), ∂u
∂xi=
I
∂uI ∂xieI.
Letbe a bounded domain. This paper is concerned with theA-Dirac equation
DAp(x,Du) = , ()
where <p<∞,Ap:×Cn→Cnpreserves the grading of the Clifford algebra and
satisfies the following conditions for some constants <a≤b<∞:
(i) The mappingx→Ap(x,ξ)is measurable for allξ∈Ckn. (ii) The mappingξ→Ap(x,ξ)is continuous for a.e.x∈. (iii) |Ap(x,ξ)| ≤b|ξ|p–.
(iv) Ap(x,ξ) –Ap(x,η),ξ–η ≥a|ξ–η|pfor allx∈withξ=η.
Definition .[] A Clifford-valued functionu∈W,p(,Ckn) fork= , , . . . ,n, is a weak solution to equation () if for allϕ∈W,p(,Ck
n) with compact support, we have ˆ
Ap(x,Du)Dϕ= . ()
In [, ], Nolder explained how quasi-linear elliptic equations
–divAp(x,∇u) = ()
arise as components of Dirac systems () and discussed some properties of the weak so-lutions to the scalar parts of equations, such as the Caccioppoli estimate and the remov-ability theorem. In [], Heinonenet al.studied the quasi-linear elliptic equations () by means of potential theory systematically. Many other mathematicians also work on prop-erties of solutions to equations (), such as the regularity, stability, convergence and so on, see [–].
This paper is organized as follows. In Section , some preliminary results about Clifford-valued functions are presented. In Section , higher integrability of weak solutions to ob-stacle problem for Clifford-valued functions are obtained. Section is concerned with the stability of the weak solutions to obstacle problem. For other works about Clifford analysis and Dirac equations, see [, , –].
2 Preliminary results
Lemma . C∞(,Ck
n)is dense in W
,p
Proof Let u=IuI(x)eI ∈W,p(,Ck
Here, we denote the Grassmann algebra=kk,dis the exterior derivative opera-tor, andd∗is the formal adjoint operator. For more details about differential form, see []. From [], we know that the linear mapping
λ:ei∧ · · · ∧eik→ei· · ·eik (k≥)
n), there exists a constantC, such that
max du , d∗u ≤C|Du|. ()
In [], Iwaniec gave the following Poincaré inequality for differential forms.
Theorem . For each u∈C∞ (,
k
),there exits a constant C(n,p), <p<∞such that ∇up≤C(n,p)dup+d∗up. ()
Proposition . For each u∈W,p(,Ck
Similar to the process of the Ponicaré inequality for differential forms in [], the fol-lowing inequalities for Clifford-valued function can be obtained.
Corollary . Let u∈W,s(,Ckn), <s<∞,then,there exists a constant C such that
In this section, we will prove the higher integrability of the weak solutions to the obstacle problemKfψ,p(,Ckn). Here, we assume that the complement ofsatisfies the measure density condition, it means that there exists a positive constantC> such that
c∩B ≥C|B|, ()
whereBis a ball inRn. In order to prove the higher integrability, we need the following
Lemma . Let Bbe a ball inRn,and let g,h∈Lp(B)be a nonnegative function
satisfy-ing
B
gpdx≤c
B g dx
p
+ B
hqdx
()
for all balls B⊂B⊂B, <p<∞.Then,for each <σ < ,p<s<p+ n–pp–nkp, we
have
σB gsdx
s
≤ n
σns( –σ) n p
B gpdx
p +
B hsdx
s
. ()
Definition . For Clifford-valued functionsa=IaI(x)eI andb=IbI(x)eI, we say thata≥binifaI(x)≥bI(x) for a.e.x∈and all ordered tupleI.
Let ψ = IψIeI be a Clifford-valued function in W,p(,Ckn), where ψI : →
[–∞, +∞). f ∈W,p(,Ck
n) is a function, which gives the boundary. We consider the
obstacle problem
Kf,p
ψ
,Ckn=v∈W,p,Ckn:v–f ∈W,p,Ckn,v≥ψa.e. ()
for Clifford-valued functions. To avoid trivialities, we always assume that the setKfψ,p(,
Ck
n) is not empty.
Definition . We say that a functionu∈Wloc,p(,Ckn) is a weak solution to the obstacle problemKfψ,p(,Ckn), if
ˆ
Ap(x,Du),Dv–Du
dx≥, ()
wheneverv∈Kψf,p(,Ckn).
Remark . Ifuis a weak solution to obstacle problem (), thenuholds for the scalar part of (),i.e.,
ˆ
ScA(x,Du)Dϕdx= . ()
Remark . We assumef ≥ψin. Indeed, denotef =IfIeI,ψ=
IψIeI. Then, since u–f ∈W,p(,Ckn),
≤(ψI–fI)+≤(u–fI)+.
Thus, we have (ψI–fI)+∈W,p() andfI= (ψI–fI)++fI≥ψIin. Letf =
IfIeI, then f ≥ψin.
Theorem . Suppose that the complement ofsatisfies the measure density condition
(),and let u∈Wloc,p(,Ck
W,p+δ(,Ck
a.e.. Thus,vis an admissible test function, and we have
≤
SinceAsatisfies (iii), (iv), combining the Hölder inequality and the Young inequality, it follows from () that
whereθ=n+pn–> . By means of condition (), we have|B∩| ≤( –C)|B|, then ()
Then, the following reverse Hölder estimate
B
Sinceis a bounded domain,can be covered by a finite number of balls such that the
previous inequality holds, then the estimates () follows immediately.
4 Stability
In this section, we will show that the weak solutions to obstacle problem are stable under some suitable assumptions.
Suppose that {pi} is a sequence such thatpi→pand for eachi= , , . . . , there is an
operatorApisatisfying (i)-(iv) forp=piand for a.e.x∈,
uniformly on compact subset ofRn. We consider the obstacle problem
Kf,pi ψ
,Ckn=v∈W,pi,Ck
n
:v–f∈W,pi
,Ckn,v≥ψa.e.. () Assume thatApipreserves the grading of the Clifford algebra and converges toAp(x,ξ) uniformly on compact subset ofRn. Suppose thatu
i∈W,pi(,Ckn) is the weak solution
to the followingKf,pi
ψ (,Ckn)-obstacle problem ˆ
Api(x,Dui),Dv–Dui
dx≥ ()
for eachv∈Kf,pi
ψ (,Ckn) andu∈W,p(,Ckn) is the weak solution to ˆ
Ap(x,Du),Dv–Du
dx≥ ()
for allv∈Kfψ,p(,Ckn).
Definition . We call a sequenceuj∈Lp(,C
n) converges weakly inLp(,Cn) touif ˆ
vujdx→ ˆ
vu dx, ()
wheneverv∈L
p p–(,C
n). Denote it byuju.
Remark . Suppose thatuiweakly converges touinLp(,Ck n), then ˆ
v,uidx→ ˆ
v,udx, ()
wheneverv∈W,p(,Ck n).
Now, we start with the main result of this section.
Theorem . Suppose that the complement ofsatisfies the measure density condition
().Let pi,Api(x,ξ),Ap(x,ξ)be defined as described above.Let the boundary value
func-tion f be in W,p+τ(,Ck
n)for someτ> .Then,there is a small numberδsuch that the
sequence{ui}of weak solutions to the obstacle problem()has a subsequence,which con-verges to the weak solution uof()in W,p+δ(,Ckn)for anyδ<δ.
Before proving Theorem ., we need the following lemmas.
Lemma . [] Suppose that a bounded open set satisfies ().Let pi> and ui ∈ W,pi(), i= , , . . . ,and suppose that pi→p and ui→u
a.e in. Ifθ ∈W,p+(), > and if ui–θ∈W,pi
()with
ˆ
∇(ui–θ) pi
dx≤M ()
Lemma . Let uibe described in Theorem..There exists a constantγ> such that
for eachγ∈[,γ],
ˆ
|Dui|p+ γ
dx≤C, ()
where i is sufficiently large and C is independent of i.
Proof By virtue of Remark ., we can usef as a test function in (), so that
ˆ
Api(x,Dui),Df –Dui
dx≥.
From the structure conditions onAiand the Hölder inequality, it follows
a ˆ
|Dui|pidx≤b
ˆ
|Dui|pidx
pi–
pi ˆ
|Df|pidx
pi .
Then we have
ˆ
|Dui|pidx≤C
ˆ
|Df|pidx. ()
By the Hölder inequality and Young’s inequality, () becomes
ˆ
|Dui|pidx≤C
||+
ˆ
|Df|p+σdx
, ()
wheniis sufficiently large andσ> . According to Theorem ., there exists a constant
γ> such that for eachγ ∈[,γ],
ˆ
|Dui|pi+γdx pi
pi+γ
≤C ˆ
|Dui|pidx+
ˆ
|Df|pi+γdx pi
pi+γ
. ()
Sincepi→p, chooseisufficiently large such thatp–γ ≤pi≤p+γ. Thenp+γ ≤p+γ. So, there exists a constant C, such that
ˆ
|Dui|p+γ dx≤C.
Proof of Theorem. Letκ=nn–. Sinceui–f ∈W,p(,Ck
n), we chooseiso large that p+γ ≤κpiandpi≤p+γ, using Lemma ., we get
ui–fp+γ ≤Cui–fκpi
≤Cui–fnpi n–pi
≤CDui–Dfpi
≤CDuip+γ +Dfp+γ
Implied by the Minkowski inequality that
uip+γ
≤ ui–fp+γ +fp+γ
≤CDuip+γ
+Dfp+γ +fp+γ
.
Then it follows from Lemma . that
ui W,p+
γ
(,Ckn)≤C. ()
Writeδ=γ,ui=IuiIeI, according to (), we haveui∈W,p+δ(,Ck
n). So, we can
ex-tract a subsequence, still denote byui
I, such that
⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩
uiIuI, inW,p+δ(),
ui
I→uI, inLp+δ(), ui
I→uI, pointwise a.e. in.
()
Then∇ui
I→ ∇uI. Letu=
IuIeI, thenui→uinLp+δ();ui→upointwise a.e. in.
Since∇ui
I∇uI, for eachj= , , . . . ,n, we have ˆ
yα ∂ui
I
∂xjdx→ ˆ
yα ∂uI
∂xj dx,
wheneveryα∈W,p(). Then,
´
yDuidx→
´
yDu dxfor eachy∈W,p(,Cn). This yieldsDuiDuinLp(,C
n).
The next stage is to extract a further subsequence, so thatDui→Dupointwise almost everywhere in.
Because ofui–f ∈W,pi
(,Ckn) andui–f →u–f, we getu–f ∈W
,pi
(,Ckn) a.e. in
, that is to say,u∈Kf,pi
ψ (,Ckn). So, ˆ
Api(x,Dui),Du–Dui
dx≥. ()
Moreover, by the convergence assumption, we obtain
Api(x,Du)→Ap(x,Du) almost everywhere in.
From this, if follows thatui→uinW,p+δ(,Ck
n) for allδ<δ.
At last, we will show thatu=u. Using Lemma (.), we getu–f ∈W,p(,Ckn), then u∈Kψf,p(,Ckn). This yields
ˆ
Ap(x,Du),Du–Du
dx≥. ()
On the other hand,
ˆ
Api(x,Dui),Du–Dui
Leti→ ∞, sinceui→uinW,p(,Ck
n), we get ˆ
Ap(x,Du),Du–Du
dx≥. ()
Combining () and (), we have
≤
ˆ
Ap(x,Du) –Ap(x,Du),Du–Du
dx≤.
It follows thatu=u. This proof is completed.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors contributed equally in this paper. They read and approved the final manuscript.
Acknowledgements
The authors would like to thank the anonymous referees for their time and thoughtful suggestions. The research is supported by the National Science Foundation of China (#11071048).
Received: 24 May 2013 Accepted: 6 August 2013 Published: 20 August 2013
References
1. Nolder, C: NonlinearA-Dirac equation. Adv. Appl. Clifford Algebras21, 429-440 (2011)
2. Nolder, C: A-harmonic equations and the Dirac operator. J. Inequal. Appl.2010, Article ID 124018 (2010) 3. Heinonen, J, Kilpeläinen, T, Martio, O: Nonlinear Potential Theory of Degenerate Elliptic Equation. Oxford University
Press, Oxford (1993)
4. Iwaniec, T, Scott, C, Stroffolini, B: Nonlinear hodge theory on manifolds with boundary. Ann. Mat. Pura Appl.CLXXVII, 37-115 (1999)
5. Iwaniec, T: The Gehring lemma. In: Quasiconformal Mappings and Analysis, pp. 181-204. Springer, Berlin (1998) 6. Li, G, Martio, O: Stability of solutions of varying degenerate elliptic equations. Indiana Univ. Math. J.47, 873-891
(1998)
7. Ran, Q, Fang, A: Stability of solution of a class of quasilinear elliptic equations. Acta Math. Appl. Sin.18, 461-470 (2002) 8. Li, G, Martio, O: Local and global integrability of gradients in obstacle problems. Ann. Acad. Sci. Fenn., Math.19, 25-34
(2004)
9. Kilpeläinen, T, Malý, J: Degenerate elliptic equations with measure data and nonlinear potentials. Ann. Sc. Norm. Super. Pisa, Cl. Sci.XIX, 591-613 (1992)
10. Chen, Q, Jost, J, Li, J, Wang, G: Regularity theorems and energy identities for Dirac-harmonic maps. Math. Z.251, 61-84 (2005)
11. Chen, Q, Jost, J, Wang, G: Nonlinear Dirac equations on Riemann surfaces. Ann. Glob. Anal. Geom.33, 253-270 (2008) 12. Nolder, C, Ryan, J:p-Dirac equations. Adv. Appl. Clifford Algebras19, 391-402 (2009)
13. Gilbert, J, Murray, M: Clifford Algebras and Dirac Operators in Harmonic Analysis. Cambridge University Press, Cambridge (1991)
14. Stroffolini, B: On weaklyA-harmonic tensors. Stud. Math.114, 289-301 (1995) doi:10.1186/1687-1847-2013-250