Over the last decade many observational results has revised the belief that diffuse clouds are simple environments (almost completely homogeneous medium, illuminated by the mean Galactic field, with physical and chemical processes dominated by the interaction with the radiation) and some questions remain unanswered. One of the most important problems is to understand the formation and
destruction mechanisms of simple molecules such as CH+ and OH, that would require formation energies much higher than those typical of diffuse clouds.
This line of research is devoted to the investigation of physical processes which, modifying the physical conditions (temperature, gas density,..) of diffuse clouds could be responsible of "anomalous" molecular formation/suppression.
In particular, the availability of the large database of far-UV spectra present in the FUSE (Far Ultraviolet Spectroscopic Explorer) Archive gives us the unique opportunity to study the excitation properties of the high rotational (J>2) levels of molecular hydrogen and to use them as a diagnostic tool of the physical properties of the medium.
We selected about 200 lines of sight from a set of OB stars observed by FUSE, showing apparent diffuse properties. We are currently carrying out the complete analysis of these spectra: in particular, applying a profile fitting method (using the free software OWENS) on the rotational lines, we are able to estimate the H2 column density in the first six (J=0-->5) rotational levels of the ground vibrational and electronic state. The following figure shows an example of the fit performed. From the relative population of H2(J=0,1), supposed to be primarily driven by thermal proton collisions, we can calculate for each line of sight the excitation temperatures T_01, which is commonly assumed as an indicator of the kinetic temperature.
From the populations of higher-J levels (determined by UV and formation pumping and radiative cascade), we can calculate other physical diagnostic tools like the excitation diagrams (see the example shown in Fig.7 ) and the ortho-to-para ratio. The large database we are analysing will allow us to study the physical conditions of diffuse clouds in a statistically significant manner. The results of this work, which have been partially part of the degree thesis of Emanuele Congiu, are currently being prepared for publication

Example of excitation diagram for J=0 through J=5. The linear fit (blue dashedline) for the T_01 corresponds to the assumed kinetic temperature derived from J=0 and J=1. The green and red lines are fit to the two couple of levels J=1,3 and J=3,5 and correspond to the excitation temperatures T_13 and T_35.