Webmaster –  Contact

.In these tables, lines are numbered in the first column. A few atomic lines, and very close H2 lines, appear in the table not in the same order as in the picture. This is because reference lines are not emitted by the same source as the H2 lines so that they may be shifted with respect to the H2 spectrum as explained in the experimental section. Sometimes a standard line appears twice in the list, which means that the line is emitted by both the low pressure discharge and the hollow cathode.

.The second column contains the relative peak intensities of H2 lines, measured from photoelectric recording but not corrected for instrumental response. Intensities are reported for H2 lines only. However no value of intensity is given for the shoulder lines indicated "sh" in the comment and also for the reversed lines, indicated "r" because for the latter lines the intensity depends strongly on the pressure in the source at the time of the recording. An intensity of zero is given to lines barely seen in the picture and the trace, for avoiding values smaller than one. It should also be remarked that, as intensities have been scaled throughout the spectrum, the figures given in the tables are rounded and may not correspond strictly to what is seen on the traces. Also the running conditions of the source may have been different at the time of the photoelectric recording from what it was when the picture was taken.

.The third and fourth columns display, respectively, the wavelength (in nm) in ascending order and the corresponding wavenumber(in cm-1).

.The fifth column indicates the abbreviated assignment of H₂ lines. In the VUV range, H₂ emission lines can be ascribed to seven band systems, namely, 2psB¹S_u⁺ -> X¹S_g⁺ (Lyman), 2ppC¹P_u -> X¹S⁺_g (Werner), as well as 3psB' and 4 psB"¹S_u⁺ -> X¹S_g⁺ and 3pp D, 4ppD' and 5ppD"¹P_u -> X¹S_g⁺ . We have also tentatively identified a few lines of the band system 6pp¹P_u -> X¹S_g⁺ . As all VUV transitions terminate on the ground state X¹S_g⁺ we have omitted X everywhere. For example, C2-7P5 designates the line J"=5 in the P-branch of the transition C¹P_u (v'=2) -> X¹S_g⁺ (v"=7).

. The last column is devoted to the identification of reference and impurity lines and may contain comments about H2 lines. For the self-reversed lines exhibiting clearly two components, the average wavenumber is repeated in the comments of the two components. Such a value is very accurate because the two components are always sharp. Finally "b" stands for blended lines, "d" for diffuse lines, and "s" for sharp unidentified lines.

Some molecular impurity lines due to N2 and CO appear on a few pictures. Their wavelengths are indicated and they are identified in the comment. At longer wavelengths, most of the unascribed lines have negligible intensities. They may be due to the presence of some impurities. At shorter wavelengths a number of unascribed lines have appreciable intensities. Most of these are certainly due to transitions from still unidentified high-levels of the states B', D, B", D', and D". These lines cannot be ascribed because the calculation does not fit with the lines predicted from absorption data well enough for a few levels of B' and D, and no calculation is yet available for B", D' and D".
For atomic lines appearing in the reference spectrum we have indicated only the wavelength unless the lines are good standards with 5 decimals. In this case we give only the literature wavelength, in the comment. All the wavelengths of reference atomic lines are extracted either from a paper by Kaufman and Edlen (1974) or from Kelly's table (1987). In the latter case the value is followed by Ke.
In a first stage of the work the lines could be identified from the values of level energies published long ago (Wilkinson 1968, Namioka 1964a and b, Dabrowski and Herzberg 1974, and Takezawa 1970), and from more recent identifications of small v', high-J levels of the Lyman band system by Dabrowski (1984). The semi-ab initio calculations of Abgrall et al. (1993a, 1993b, 1993c, and 1994) give emission probabilities and wavenumbers of most lines to an accuracy of about 0.5 cm -1. Such a calculation was necessary, in order to assign unambiguously many high-J lines. In that way about 96% of the lines have at least one assignment.
The absence of most lines missing in the tables of Abgrall et al. (1993b, 1993c, and 1994) is well explained either by radiation trapping or a weak emission probability. Typical cases have been illustrated by Abgrall et al. (1993a). Most mistakes, but probably not all of them, have been tracked down by several error tests included in the file handling programs used for editing the tables of Abgrall et al. (1993b, 1993c, and 1994).
It has to be pointed out that, due to interactive work with theoreticians, the present atlas is more than a simple catalog of spectral lines, as 96% of molecular lines are identified, most of them for the first time. It is hoped that this book will be useful to scientists in many fields and particularly to: . astrophysicists who study the interstellar medium as well as those who are receiving, from spacecrafts, VUV spectra emitted by planetary atmospheres where H2 is the major constituent, . physicists who study emission of H2 excited by electron impact in connection with planetary atmosphere processes, . laser physicists, who need a very good knowledge of molecular line wavelengths, . theoreticians for testing any improvement of the model.