Some new estimates for generalized fractional integrals associated with operators on Morrey spaces

Hua Wang · 2026

Read Some new estimates for generalized fractional integrals associated with operators on Morrey spaces by Hua Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-19372.txt.

Introduction

Wang proves new bounds for generalized fractional integrals tied to operators with Gaussian heat-kernel spread. For a fractional power between zero and n, he shows the map sends a Morrey space into a space of bounded mean oscillation tied to the map. The vanishing versions match too. At the endpoint exponent the map runs from weak Lebesgue space into the same oscillation space. Proofs rest on sharp kernel estimates for the fractional map and a damped partner.

Catalog record

CREATORHua Wang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE63292
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.19372

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Some new estimates for generalized fractional integrals associated with operators on Morrey spaces Hua Wang Abstract . Let ℒ \mathcal{L} be the infinitesimal generator of an analytic semigroup { e − t ​ ℒ : t > 0 } \big\{e^{-t\mathcal{L}}:t>0\big\} on L 2 ​ ( ℝ n ) L^{2}(\mathbb{R}^{n}) with Gaussian upper bounds, and suppose that ℒ \mathcal{L} has a bounded holomorphic functional calculus on L 2 ​ ( ℝ n ) L^{2}(\mathbb{R}^{n}) . For given 0 < α < n 0<\alpha<n , let ℒ − α / 2 \mathcal{L}^{-\alpha/2} be the generalized fractional integral associated with ℒ \mathcal{L} , which is given by ℒ − α / 2 ( f ) ( x ) := 1 Γ ⁡ ( α / 2 ) ∫ 0 + ∞ e − t ​ ℒ ( f ) ( x ) t α / 2 − 1 d t , \mathcal{L}^{-\alpha/2}(f)(x):=\frac{1}{\Gamma(\alpha/2)}\int_{0}^{+\infty}e^{-t\mathcal{L}}(f)(x)t^{\alpha/2-1}dt, where Γ ⁡ ( ⋅ ) \Gamma(\cdot) is the usual gamma function. In the limiting Sobolev case λ = n − α ​ p \lambda=n-\alpha p and 1 ≤ p < n / α 1\leq p<n/{\alpha} , the author proves that the operator ℒ − α / 2 \mathcal{L}^{-\alpha/2} is bounded from the Morrey space M p , λ ​ ( ℝ n ) M^{p,\lambda}(\mathbb{R}^{n}) into BMO ℒ ​ ( ℝ n ) \mathrm{BMO}_{\mathcal{L}}(\mathbb{R}^{n}) , and is bounded from the vanishing Morrey space V ​ M p , λ ​ ( ℝ n ) VM^{p,\lambda}(\mathbb{R}^{n}) into VMO ℒ ​ ( ℝ n ) \mathrm{VMO}_{\mathcal{L}}(\mathbb{R}^{n}) , where BMO ℒ ​ ( ℝ n ) \mathrm{BMO}_{\mathcal{L}}(\mathbb{R}^{n}) and VMO ℒ ​ ( ℝ n ) \mathrm{VMO}_{\mathcal{L}}(\mathbb{R}^{n}) are the spaces of bounded mean oscillation and vanishing mean oscillation associated with the operator ℒ \mathcal{L} , respectively. As a consequence, the author obtains that the operator ℒ − α / 2 \mathcal{L}^{-\alpha/2} is bounded from L p , ∞ ​ ( ℝ n ) L^{p,\infty}(\mathbb{R}^{n}) into BMO ℒ ​ ( ℝ n ) \mathrm{BMO}_{\mathcal{L}}(\mathbb{R}^{n}) when p = n / α p=n/{\alpha} and 0 < α < n 0<\alpha<n . The proofs are based on pointwise kernel estimates of the operators ℒ − α / 2 \mathcal{L}^{-\alpha/2} and ( I − e − t ​ ℒ ) ℒ − α / 2 (I-e^{-t\mathcal{L}})\mathcal{L}^{-\alpha/2} for 0 < α < n 0<\alpha<n . 2020 Mathematics Subject Classification Primary 42B20, 42B35; Secondary 46E30, 47G10 1. Introduction and preliminaries Let \mathbb{R}^{n} be the n -dimensional Euclidean space endowed with the Lebesgue measure dx and the Euclidean norm |\cdot| . Let \mathcal{L} be the infinitesimal generator of an analytic semigroup \big\{e^{-t\mathcal{L}}:t>0\big\} on L^{2}(\mathbb{R}^{n}) with Gaussian upper bounds on its heat kernel, and suppose that \mathcal{L} has a bounded holomorphic functional calculus on L^{2}(\mathbb{R}^{n}) . Then \mathcal{L} is the linear operator on L^{2}(\mathbb{R}^{n}) which generates an analytic semigroup \big\{e^{-t\mathcal{L}}:t>0\big\} with kernel \mathcal{P}_{t}(x,y) satisfying e^{-t\mathcal{L}}(f)(x):=\int_{\mathbb{R}^{n}}\mathcal{P}_{t}(x,y)f(y)\,dy,\quad t>0, and there exist two positive constants C and A such that \big|\mathcal{P}_{t}(x,y)\big|\leq\frac{C}{t^{n/2}}\cdot e^{-A\frac{|x-y|^{2}}{t}} (1.1) holds for all x,y\in\mathbb{R}^{n} and all t>0 . For any 0<\alpha<n , the generalized fractional integral \mathcal{L}^{-\alpha/2} associated with the operator \mathcal{L} is defined by \mathcal{L}^{-\alpha/2}(f)(x):=\frac{1}{\Gamma(\alpha/2)}\int_{0}^{+\infty}e^{-t\mathcal{L}}(f)(x)t^{\alpha/2-1}dt,\quad x\in\mathbb{R}^{n}. (1.2) Let \Delta be the Laplacian operator on \mathbb{R}^{n} , that is, \Delta:=\frac{\partial^{2}}{\partial x_{1}^{2}}+\frac{\partial^{2}}{\partial x_{2}^{2}}+\cdots+\frac{\partial^{2}}{\partial x_{n}^{2}}. Note that if \mathcal{L}=-\Delta is the Laplacian operator on \mathbb{R}^{n} , then \mathcal{L}^{-\alpha/2} is exactly the classical fractional integral(or Riesz potential operator) \mathcal{I}_{\alpha} of order \alpha ( 0<\alpha<n ), which is given by \mathcal{I}_{\alpha}(f)(x):=\frac{1}{\gamma(\alpha)}\int_{\mathbb{R}^{n}}\frac{f(y)}{|x-y|^{n-\alpha}}dy,\quad x\in\mathbb{R}^{n}, where \gamma(\alpha):=\frac{2^{\alpha}\pi^{n/2}\Gamma(\alpha/2)}{\Gamma({(n-\alpha)}/2)} and \Gamma(\cdot) being the usual gamma function. Example 1.1 . The Gaussian upper bound ( 1.1 ) is satisfied by a large class of differential operators. Some of them are listed below: (1) Let 0\leq V(x)\in L^{1}_{\mathrm{loc}}(\mathbb{R}^{n}) . The Schrödinger operator with nonnegative potential V(x) is defined by \mathcal{L}_{1}=-\Delta+V\quad\mbox{on}~~\mathbb{R}^{n}. The operator \mathcal{L}_{1} is a self-adjoint positive definite operator. Hence it has a bounded functional calculus on L^{2}(\mathbb{R}^{n}) . By the well-known Feynman–Kac formula, we can see that the kernel \mathcal{P}_{t}(x,y) of the semigroup \big\{e^{-t\mathcal{L}_{1}}:t>0\big\} satisfies the following pointwise estimate 0\leq\mathcal{P}_{t}(x,y)\leq\frac{1}{(4\pi t)^{n/2}}e^{-\frac{|x-y|^{2}}{4t}}. That is, the Gaussian upper bound ( 1.1 ) is satisfied. For some more interesting results on the generalized fractional integral \mathcal{L}_{1}^{-\alpha/2} generated by the operator \mathcal{L}_{1} , we refer the reader to [ 3 ] , [ 6 ] , [ 28 ] , [ 30 ] and the references therein. (2) Let \mathcal{A}=\mathcal{A}(x) be an n\times n matrix of complex L^{\infty} -coefficients defined on \mathbb{R}^{n} . For \xi=(\xi_{1},\xi_{2},\dots,\xi_{n})\in\mathbb{C}^{n} , we denote its complex conjugate (\bar{\xi_{1}},\bar{\xi_{2}},\dots,\bar{\xi_{n}}) by \bar{\xi} . Assume that this matrix satifies the following ellipticity (or “accretivity”) condition: \lambda|\xi|^{2}\leq\mathrm{Re}\,\mathcal{A}\xi\cdot\bar{\xi}\quad\mbox{and}\quad|\mathcal{A}\xi\cdot\bar{\zeta}|\leq\Lambda|\xi||\zeta| for all \xi,\zeta\in\mathbb{C}^{n} and for some constants \lambda,\Lambda such that 0<\lambda\leq\Lambda<\infty . Here we use the inner product notation \xi\cdot\bar{\zeta}=\xi_{1}\bar{\zeta_{1}}+\xi_{2}\bar{\zeta_{2}}+\cdots+\xi_{n}\bar{\zeta_{n}}. Then we define the second-order divergence form operator as \mathcal{L}_{2}=-\mathrm{div}(\mathcal{A}\nabla)\quad\mbox{on}~~\mathbb{R}^{n}, which we interpret in the standard weak sense of maximal accretive operators via a sesquilinear form. The maximal accretivity condition ensures the existence of an analytic contraction semigroup on L^{2}(\mathbb{R}^{n}) generated by -\mathcal{L}_{2} . In this setting, the complex elliptic operator \mathcal{L}_{2} also has a bounded holomorphic functional calculus on L^{2}(\mathbb{R}^{n}) . It is known that when \mathcal{A} has real entries, or when n=1,2 in the case of complex entries, the operator \mathcal{L}_{2} generates an analytic semigroup \big\{e^{-t\mathcal{L}_{2}}:t>0\big\} with heat kernel satisfying the Gaussian upper bound ( 1.1 )(see, for example, [ 12 ] and [ 13 ] ). In general, when n\geq 3 in the case of complex entries, the property ( 1.1 ) on the heat kernel does not hold. For some more results on the generalized fractional integral \mathcal{L}_{2}^{-\alpha/2} generated by the operator \mathcal{L}_{2} , we refer the reader to [ 4 ] , [ 5 ] , [ 9 ] and the references therein. Since the semigroup \big\{e^{-t\mathcal{L}}:t>0\big\} has a kernel \mathcal{P}_{t}(x,y) which satisfies the Gaussian upper bound ( 1.1 ), it is easy to check that there exists a constant C>0 such that \big|\mathcal{L}^{-\alpha/2}(f)(x)\big|\leq C\cdot\mathcal{I}_{\alpha}(|f|)(x) (1.3) holds for all x\in\mathbb{R}^{n} . In fact, if we denote the kernel of \mathcal{L}^{-\alpha/2} by \mathcal{K}_{\alpha}(x,y) , then it follows from ( 1.2 ) and Fubini’s theorem that \begin{split}\int_{\mathbb{R}^{n}}\mathcal{K}_{\alpha}(x,y)f(y)\,dy&=\mathcal{L}^{-\alpha/2}(f)(x)\\ &=\frac{1}{\Gamma(\alpha/2)}\int_{0}^{+\infty}e^{-t\mathcal{L}}(f)(x)t^{\alpha/2-1}dt\\ &=\int_{0}^{+\infty}\bigg\{\frac{1}{\Gamma(\alpha/2)}\int_{\mathbb{R}^{n}}\mathcal{P}_{t}(x,y)f(y)\,dy\bigg\}t^{\alpha/2-1}dt\\ &=\int_{\mathbb{R}^{n}}\bigg\{\frac{1}{\Gamma(\alpha/2)}\int_{0}^{+\infty}\mathcal{P}_{t}(x,y)t^{\alpha/2-1}dt\bigg\}f(y)\,dy.\end{split} Then we have \mathcal{K}_{\alpha}(x,y)=\frac{1}{\Gamma(\alpha/2)}\int_{0}^{+\infty}\mathcal{P}_{t}(x,y)t^{\alpha/2-1}dt, (1.4) where \mathcal{P}_{t}(x,y) is the kernel of e^{-t\mathcal{L}} . Thus, by using the Gaussian upper bound ( 1.1 ) and the expression ( 1.4 ), we can deduce that for any 0<\alpha<n , \displaystyle\big|\mathcal{K}_{\alpha}(x,y)\big| \displaystyle\leq\frac{1}{\Gamma(\alpha/2)}\int_{0}^{+\infty}\big|\mathcal{P}_{t}(x,y)\big|t^{\alpha/2-1}dt \displaystyle\leq C\cdot\int_{0}^{+\infty}e^{-A\frac{|x-y|^{2}}{t}}\cdot t^{\alpha/2-n/2-1}dt \displaystyle\leq C\cdot\frac{1}{|x-y|^{n-\alpha}}\int_{0}^{+\infty}e^{-\nu}\cdot\nu^{n/2-\alpha/2-1}d\nu \displaystyle\leq C\cdot\frac{1}{|x-y|^{n-\alpha}}. (1.5) This proves ( 1.3 ) with C>0 independent of f (see [ 13 ] and [ 19 ] ). Recall that, for any given 1\leq p<\infty , the space L^{p}(\mathbb{R}^{n}) is defined as the set of all integrable functions f on \mathbb{R}^{n} such that \|f\|_{L^{p}}:=\bigg(\int_{\mathbb{R}^{n}}|f(x)|^{p}dx\bigg)^{1/p}<+\infty, and the weak space L^{p,\infty}(\mathbb{R}^{n}) is defined as the set of all measurable functions f on \mathbb{R}^{n} such that \|f\|_{L^{p,\infty}}:=\sup_{\lambda>0}\lambda\cdot m\big(\big\{x\in\mathbb{R}^{n}:|f(x)|>\lambda\big\}\big)^{1/p}<+\infty. In the sequel, for a Lebesgue measurable set E\subset\mathbb{R}^{n} , the n -dimensional Lebesgue measure of E is denoted by m(E) . For the case p=\infty , L^{\infty,\infty}(\mathbb{R}^{n}) will be taken to mean L^{\infty}(\mathbb{R}^{n}) , which is defined as the set of all measurable functions f on \mathbb{R}^{n} such that \|f\|_{L^{\infty}}:=\underset{x\in\mathbb{R}^{n}}{\mbox{ess\,sup}}\,|f(x)|<+\infty. A real-valued locally integrable function f on \mathbb{R}^{n} is said to be in \mathrm{BMO}(\mathbb{R}^{n}) , the space of bounded mean oscillation, if \|f\|_{\mathrm{BMO}}:=\sup_{\mathcal{B}\subset\mathbb{R}^{n}}\frac{1}{m(\mathcal{B})}\int_{\mathcal{B}}|f(x)-f_{\mathcal{B}}|\,dx<+\infty, (1.6) where the supremum is taken over all balls \mathcal{B} contained

Read the complete text