Investigation of magnetically sensitive FeH lines

TL;DR

This study investigates how FeH molecular lines in M-dwarf stars respond to magnetic fields, analyzing dependencies on quantum numbers and comparing theoretical predictions with high-resolution observations.

Contribution

It provides new insights into the magnetic sensitivity of FeH lines, especially regarding the influence of the projection of angular momentum, {\Omega}, and compares computed and observed spectra.

Findings

Lines with high {\Omega} show strong magnetic sensitivity.

No clear correlation with branch or J was observed.

Computed spectra generally agree with observations, but some lines show discrepancies.

Abstract

M-type stars exhibit strong magnetic fields towards decreasing effective temperatures. The measurement of these fields is complicated due to missing indicators. Molecular FeH lines provide an excellent means to determine magnetic field strengths from the Zeeman broadening of magnetically sensitive lines. Our aim is the investigation of possible dependencies of the amount of sensitivity to magnetic fields from rotational quantum number, branch, and the projection of the total angular momentum onto the internuclear axis ({\Omega}). We also compare results from com-putations with those from observations. We use high resolution CRIRES spectra of the two M dwarfs GJ1002 (M5.5 inactive) and GJ1224 (M4.5 active). Individual lines are fitted by Gaussians and the obtained line depths and widths from the active and inactive star can be compared with each other. In this way, magnetically sensitive lines can be detected. For test purposes, we do the same with computed spectra of FeH. One with zero magnetic field and the other with a 2 kG magnetic field vector used in the disc integration (i.e. pure radial at the disc center). We found, in agreement with theory, that lines with high {\Omega} show strong sensitivity to magnetic fields. No obvious correlation with branch or J was found, which was also expected for lines formed in intermediate Hund's case. The computations agreed in general well with the observations, but in many cases the individual splitting of certain lines can be very different to observations.