1 Introduction

Astrophysical and laboratory applications often require large datasets that are complete and accurate for comprehensive model calculations of opacities (Seaton et al.1994, the Opacity Project Team 1995, 1996), radiative forces (e.g. Seaton 1997; Hui-Bon-Hoa & Alecian 1998; Seaton 1999), radiation transport in high-density fusion plasmas, etc. The Opacity Project (OP) (The Opacity Project 1995, 1996; Seaton et al. 1994) produced large datasets of transition probabilities for most astrophysically abundant atomic systems in the close coupling approximation using the powerful R-matrix method from atomic collision theory (Burke et al.1971; Seaton 1987; Berrington et al.1987). However, the calculations were carried out in LS coupling and the A-values were obtained neglecting relativistic fine-structure. The LS multiplets may be divided into fine-structure components using algebraic transformations. This has been done for a number of atoms and ions using the OP data (or similar non-relativistic calculations), including iron ions such as Fe II (Nahar 1995), Fe III (Nahar & Pradhan 1996), and Fe XIII (Nahar 1999). However, for such complex and heavy ions the neglect of relativistic effects may lead to a significant lack of precision, especially for weak transitions.

As an extension of the OP to include relativistic effects, the present Iron Project (IP) (Hummer et al.1993) employs relativistic extensions of the R-matrix codes in the Breit-Pauli approximation (Scott & Burke 1980; Scott & Taylor 1982; Berrington et al. 1995) to compute radiative and collisional atomic parameters. Recently, several relativistic calculations of transition probabilities have been carried out using the Breit-Pauli R-matrix method (BPRM); e.g. for Fe XXV and Fe XXIV (Nahar & Pradhan 1999), C III (Berrington et al.1998), Fe XXIII (Ramírez et al. 2000). These calculations produced highly accurate oscillator strengths for most transitions considered, within a few percent of experimental data or other accurate theoretical calculations (where available).

However, in these relatively simple atomic systems the electron correlation effects are weak and the configuration-interaction (CI, in the atomic structure sense) is easier to account for than in the more complex ions such as the low ionization stages of iron. In the present report we present the results of a large-scale BPRM calculation for one such ion, Fe V, and discuss the accuracy and completeness of the calculated data. An earlier work (Nahar & Pradhan 2000) has decribed certain important aspects of these calculations, in particular the difficulty with the identification of levels and completeness of fine-structure components within the LS multiplets. The general aim of the present work is two-fold: (i) to extend the IP work to the calculation of relativistic transition probabilities for the complex low-z iron ions, and (ii) to provide a detailed description of the extensive data tables that should be essentially complete for most applications.