A generalization of Fueter's theorem (Q867482)

From MaRDI portal
scientific article
Language Label Description Also known as
English
A generalization of Fueter's theorem
scientific article

    Statements

    A generalization of Fueter's theorem (English)
    0 references
    0 references
    0 references
    15 February 2007
    0 references
    There is obtained a far-reaching generalization of Fueter's theorem on the construction of monogenic quaternionic functions starting from a holomorphic function in the upper half-plane. The following notation is used. \(A\) given \(\mathbb R^m\) is split as \(\mathbb R^m= \sum_{s=1}^d \bigoplus \mathbb R^{p_s}, \quad \sum_{s=1}^d p_s= m\), yielding \[ \mathbf x= \sum_{s=1}^d \mathbf x^{(s)}, \quad \mathbf x^{(s)}= \sum_{s=1}^{P_s} e_j^{(s)} x_j^{(s)} \] and accordingly \[ \partial_{\mathbf x}= \sum_{s=1}^d \partial_{\mathbf x^{(s)}}, \quad \partial_{\mathbf x^{(s)}}= \sum_{j=1}^{p_s} e_j^{(s)} \partial_{x_j^{(s)}} \] where the notations \(e^{(s)}_j\) and \(x^{(s)}_j\) have the usual in Clifford analysis meaning. Set \(r_s=| \mathbf x^{(s)}| , \quad \underline{\omega}_s = \frac{\mathbf x^{(s)}}{r_s},\quad s=1,2,\dots, d\). Denote by \(\mathbb R_{0,d}\) the \(2^d\)-dimensional Clifford algebra generated by the elements \(E_s, \;s=1,2,\dots,d\), where \[ E_s^2=-1,\quad s=1,\dots, d, \qquad E_s E_{s'}+ E_{s'} E_s=0,\quad s\neq s', \] and let \(\partial_{\underline{r}} =\sum_{s=1}^d E_s \partial_{r_s}\) be the associated Dirac operator. Let \[ G_l(r_1, \dots, r_d) = \sum_{| B| =l} G_{l,B}(r_1,\dots, r_d) E_B \] where \(l \;(1\leq l \leq d)\) is fixed integer, \(E_B= E_{\beta_1}E_{\beta_2}\dots E_{\beta_l}\) and \[ B= \{ \beta_1,\beta_2 ,\dots, \beta_l\}\subset \{1,2,\dots, d\}, \;\beta_1< \cdots < \beta_l, \] be an \(l\)-vector valued function. It is assumed that the function \(G_l\) is left monogenic with respect to the Dirac operator \(\partial_{\underline{r}}\), i.e., \(\partial_{\underline{r}}G_l =0\) or equivalently \[ \sum_{B}\sum_{s=1}^l (-1)^s\partial_{r_{\beta_s}}G_{l,B} E_{B\setminus \{\beta_s\}}=0 \] \[ \sum_{B}\;\sum_{s=1, \;s\notin B }^d \partial_{r_s}G_{l,B} E_s E_B =0. \] Then the following theorem holds. Theorem 1.1: Let \(G_l\) be as above. Assume \(p_s \;(s=1,2,\dots,d)\) to be odd and let \(\underline{\omega}_B= \underline{\omega}_{\beta_1}\underline{\omega}_{\beta_2}\dots \underline{\omega}_{\beta_l}\), \[ \mathbf{P}_k(\mathbf x)=\prod_{s=1}^d P_{k_s}( \mathbf x^{(s)}) \] with \(k=\sum_{s=1}^d k_s\) and \(P_{k_s}( \mathbf x^{(s)})\) a homogeneous left monogenic polynomial of degree \(k_s\) in \(\mathbb R^{p_s}\) with values in \(\mathbb R_{0,p_s}\). Then \[ \Delta_{\mathbf x}^{k+\frac{m-d}{2}} \left[ \left( \sum_{| B| =l} G_{l,B} (r_1, \dots, r_d)\underline{\omega}_{B} \right)\mathbf{P}_k(\mathbf x) \right] \] is left monogenic with respect to \(\partial_{\underline{x}}\).
    0 references
    0 references
    Dirac operator
    0 references
    monogenic functions
    0 references
    0 references