Compensated compactness, paracommutaors, and Hardy spaces (Q1376553): Difference between revisions

Let \(B_1: \mathbb{R}^n\times \mathbb{R}^{N_1}\rightharpoonup \mathbb{R}^{m_1}\), \(B_2: \mathbb{R}^n\times \mathbb{R}^{N_2}\rightharpoonup \mathbb{R}^{m_2}\) and \(Q: \mathbb{R}^{m_2}\rightharpoonup \mathbb{R}^{m_1}\) be bilinear forms which are related as follows: if \(\mu\) and satisfy \(B_1(,\mu)= 0\) and \(B_2(,)= 0\) for some \(\neq 0\), then \(\mu Q= 0\). Suppose \(p^{-1}+ q^{-1}= 1\). Coifman, Lions, Meyer and Semmes proved that, if \(u\in L^p(\mathbb{R}^n)\) and \(v\in L^q(\mathbb{R}^n)\), and the first-order systems \(B_1(D,u)= 0\), \(B_2(D,v)= 0\) hold, then \(uQv\) belongs to the Hardy space \(H^1(\mathbb{R}^n)\), provided that both (i) \(p= q=2\), and (ii) the ranks of the linear maps \(B_j(,\cdot): \mathbb{R}^{N_j}\rightharpoonup \mathbb{R}^{m_1}\) are constant. We apply the theory of paracommutators to show that this result remains valid when only one of the hypotheses (i), (ii) is postulated. The removal of the constant-rank condition when \(p=q=2\) involves the use of a deep result of Lojasiewicz from singularity theory.