1 Erratum to: J Stat Phys (2014) 154:866–876 DOI 10.1007/s10955-014-0915-0
A sign misprint in the statement of Proposition 3.1 in [1] has a consequence in the statement of the main Theorem 1.1 in [1]. We write a short note in which we present the correct statements.
The correct version of Proposition 3.1 in [1] is the following one.
Proposition 0.1
Let \(A:{\mathbb R}^3\rightarrow {\mathbb R}^3\). For any \(\phi =\phi (x)\in \mathcal C^{\infty }_0({\mathbb R}^3;{\mathbb C}^2)\), the following identity holds
where \(r:=|x|\), \(\partial _r^A:=\frac{x}{r}\cdot \nabla _A\), and \(A_r:=A\cdot \frac{x}{r}\).
The main difference between Proposition 3.1 in [1] and the proposition above is the negative sign in from of the last term in (0.1).
We also underline that the identity (0.1) follows by the identities (3.7) and (3.11) in [1], whose proof given in [1], is correct.
In order to present a correct version of Theorem 1.1 in [1] we first need the following version of Proposition 2.1 in [1].
Proposition 0.2
Let \(B:{\mathbb R}^3\setminus \{0\}\rightarrow {\mathbb R}^3\), and assume that there exists a vector potential \(A:{\mathbb R}^3\rightarrow {\mathbb R}^3\) such that the following holds
for all \(x\in {\mathbb R}^3\setminus \{0\}\), \(\lambda \ne 0\). Then we have
for \(\in {\mathbb R}^3\setminus \{0\}\).
Proof
The proof is quite immediate. Due to the first condition in (0.2), one has \(x\wedge \nabla (A\cdot x/r)=0\). Moreover, since \(A\) is homogeneous of degree \(-1\), then \(x\wedge A\) is homogeneous of degree \(0\), hence \(\partial _r(x\wedge A)\equiv 0\). \(\square \)
We can now state the new and correct version of Theorem 1.1 in [1].
Theorem 0.1
Let \(B:{\mathbb R}^3\setminus \{0\}\rightarrow {\mathbb R}^3\), and assume that there exists a vector potential \(A:{\mathbb R}^3\rightarrow {\mathbb R}^3\) such that the following holds
for all \(x\in {\mathbb R}^3\setminus \{0\}\), \(\lambda \ne 0\). Then, for any \(\phi =\phi (x)\in \mathcal C^{\infty }_0({\mathbb R}^3;{\mathbb C}^2)\), the following inequality holds
where \(\mu _1=\mu _1(A)=\inf \left\{ \text {spec} (\sigma \cdot {L}_{A}+1) \cap [0, \infty )\right\} \).
Proof
The proof is an immediate corollary of Propositions 0.1 and 0.2, and the usual radial Hardy inequality. \(\square \)
References
Fanelli, L., Vega, L., Visciglia, N.: Relativistic Hardy inequalities in homogeneous magnetic fields. J. Stat. Phys. 154, 866–876 (2014)
Author information
Authors and Affiliations
Corresponding author
Additional information
The online version of the original article can be found under doi:10.1007/s10955-014-0915-0.
Rights and permissions
About this article
Cite this article
Fanelli, L., Vega, L. & Visciglia, N. Erratum to: Relativistic Hardy Inequalities in Magnetic Fields. J Stat Phys 158, 1413–1414 (2015). https://doi.org/10.1007/s10955-014-1171-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10955-014-1171-z