1 Introduction

The pump mechanism of OH masers in star forming regions (e.g. W3(OH)) has been a field of intensive research in the last decade. There are several rival models for this pump mechanisms under discussion (Andresen 1985; Piehler & Kegel 1989; Cesaroni & Walmsley 1991; Gray et al. 1992; Elitzur 1996). A proper understanding of these masers would deliver detailed information about the physical conditions in these regions, e.g. grain temperature and grain density, magnetic fields, OH and $\rm H_2O$ densities, radiation fields (Reid 1993; Gray & Field 1995; Pavlakis & Kylafis 1996; Thissen et al. 1999). OH masers could then be used as a diagnostics tool for regions of star formation.

We assume that photodissociation of $\rm H_2O$ in the first absorption band (135-190 nm) plays a central role in the pumping mechanism for OH masers (Andresen et al. 1985; Thissen et al. 1999).

This work is one step in a series of laboratory experiments to measure the nascent OH population out of photodissociation of $\rm H_2O$. It was shown by Andresen et al. 1985 that photodissociation of rotationally cold $\rm H_2O$ at 157 nm leads to a population inversion between the $\Lambda$-doublets of the OH-fragment. Wurps et al. (1996) reported a population inversion between the $\Lambda$-doublets for the OH $^2\Pi _{3/2}(J=7/2)$ state of 1.8:1. Wurps measured the changing of the $\Lambda$-doublet population for the OH $^2\Pi _{3/2}(J=7/2)$ state by microwave stimulation. Linebroadening mechanisms for the hyperfine transitions were understood quantitatively.

This paper deals with the question whether photodissociation of $\rm H_2O$ yields a non statistic population of the hyperfine levels of each $\Lambda$-doublet to explain satellite line maser emission. However, a statistical population is found within the error limits of the experiment.