Mon. Not. R. Astron. Soc. 000, 1–7 (2008) Printed 17 June 2018 (MN LATEX style file v2.2) Centrosymmetric molecules as possible carriers of diffuse interstellar bands ⋆ M. Ka´zmierczak,1 † M. R. Schmidt,2 ‡ G. A. Galazutdinov,3 § F. A. Musaev,4 Y. Betelesky,5 ¶ J. Kre lowski1 k 1 Centre for Astronomy, Nicolaus Copernicus University, Gagarina 11, 87-100 Toru´n, Poland 2 Nicolaus Copernicus Astronomical Center, Rabia´nska 8, 87-100 Toru´n, Poland 3 Instituto de Astronomia, Universidad Catolica del Norte, Angamos 0610, Antofagasta, Chile 4 International Centre for Astronomical and Medico-Ecological Research, Terskol, Russia 5 European Southern Observatory, Karl-Schwarzschild-Strasse 2, 85748 Garching, Germany Accepted... Received ... in original form ... ABSTRACT C 1 1 + This paper presents a new data with interstellar 2 (Phillips bands A Πu − X Σg ) which were observed with ESO/UVES spectrograph. We determined interstellar col- umn densities and excitation temperatures of C2 for nine Galactic lines. For seven of them C2 have never been observed before, so in that case still small sample of in- terstellar clouds (26 lines of sight) where a detailed analysis of excitation of C2 was made, increased significantly. This paper is a continuation of previous works where interstellar molecules (C2 and diffuse interstellar bands) were analysed. Since a sam- ple of interstellar clouds with C2 risen we can show that the width and shape of some DIB’s profile (6196 A,˚ 5797 A)˚ apparently depend on the gas kinetic and rotational temperature of C2, being broader for its higher values. There are also DIBs (4964 A,˚ 5850 A)˚ for which that effect does not exist. Key words: ISM: lines and bands – molecules 1 INTRODUCTION tic probes of conditions in molecular clouds that produce the interstellar absorption lines, in contrast to polar molecules, C2 is the simplest multi-carbon species. As a homonu- such as CH or CN, where usually only a few absorption arXiv:1009.5194v1 [astro-ph.GA] 27 Sep 2010 clear diatomic molecule, has a negligible dipole moment lines from the lowest rotational levels are observed. C2 is a and hence radiative cooling of the excited rotational lev- useful tool to determine physical conditions (temperatures els may go only through the slow quadrupole transi- and densities) in interstellar clouds. tions (van Dishoeck & Black 1982). The rotational levels are pumped by the galactic interstellar radiation field and ex- Moreover, C2 abundances may give information on the cited effectively above the gas kinetic temperature. Lines chemistry of the intervening clouds, especially on the path- of the diatomic carbon from a long-lived ground state rota- way to the formation of long chain carbon molecules which tional levels are measurable and can be the sensitive diagnos- may be connected with carriers of diffuse interstellar bands (DIBs) (Douglas 1977; Maier et al. 2006). Diffuse interstel- lar bands (DIBs) were first time observed in interstellar ⋆ medium by Heger (1922). At present we know 414 DIBs Based on observations made with ESO Telescopes at the (Hobbs et al. 2009), but none of them has been identified Paranal Observatory under programme IDs 266.D-5655(A), 67.C- yet in spite of being the subject of much observational and 0281(A), 71.C-0513(C), 67.D-0439(A), 082.C-0566(A) and La Silla under programme IDs 078.C-0403(A), 076.C-0164(A) and theoretical research. 073.C-0337(A); and also observations made with the 1.8m tele- Useful information allowing identification of diffuse in- scope in South Korea and 2m telescope in International Centre for terstellar bands may come from an analysis of their pro- Astronomical and Medico-Ecological Research, Terskol, Russia. files. They can vary (Herbig 1975; Westerlund & Krelowski † e-mail: [email protected] 1988; Galazutdinov et al. 2002) or have some substruc- ‡ e-mail: [email protected] § email: [email protected] tures inside the profiles (Sarre et al. 1995) not only be- ¶ email: [email protected] cause of multi interstellar clouds toward one star causing k e-mail: [email protected] the Doppler splitting which can likely modify the profiles c 2008 RAS 2 M. Ka´zmierczak et al. of DIBs (Herbig & Soderblom 1982). Ka´zmierczak et al. Table 1. Basic data for the programme stars (2009) showed that there is a relation between the profile widths of strong diffuse interstellar band at 6196 A˚ and the object name Sp/L E(B-V) excitation temperatures of C2. The 6196 A˚ carrier could be a centrosymmetric molecule, whose spectral features become HD 115842 B0.5Ia 0.49 broader as their rotational temperatures increase. HD 136239 CZ Cir B1.5Ia 1.11 HD 148379 B2Iab 0.71 The goal of this paper is at first to increase sample of HD 149757 ζ Oph O9.5V 0.28 interstellar clouds where a detailed analysis of excitation of HD 151932 WN7A 0.35 C2 was made (since that time we had known 26 objects with HD 152236 B1Iape 0.66 interstellar C2, 24 lines of sight - see (Sonnentrucker et al. HD 154368 V1074 Sco O9.5Iab 0.78 2007) - Table 13 plus 2 new objects in (Ka´zmierczak et al. HD 154445 B1V 0.35 2010)). HD 170740 B2V 0.45 The second goal of this paper is to analyse other dif- fuse interstellar bands in comparison to excitation tem- peratures of C2. We found that some of DIB profiles 2.2 Results (6196 A˚ (Ka´zmierczak et al. 2009), 5797 A˚ (this paper)) widths apparently depend on the C2 temperature, being We have identified absorption lines of the (1,0), (2,0), 1 1 + broader for its higher values. There are also DIBs (4964 A,˚ (3,0) bands of the C2 Phillips system (A Πu − X Σg ) 5850 A)˚ for which that effect does not exist. Although effect (P, Q, R branches in bands (1,0) 10133 - 10262 A,˚ (2,0) is very subtle, the difference is evident. 8750 - 8849 A,˚ (3,0) 7714 - 7793 A).˚ The equivalent widths In Section 2 and 3, we describe C2 and analysed dif- with errors of all measured interstellar lines of C2 towards fuse interstellar bands, respectively. General discussion and our program stars are given in Tables 2-4. summary of our conclusions are given in Section 4. For the optically thin case (when the absorption lines are on the linear part of the curve of growth) column density of a rotational level J” can be derived from the equivalent width Wλ [mA]˚ of the single absorption line using the rela- 2 C2 tionship (Frisch 1972) 2.1 The observational data × 17 Wλ Ncol = 1.13 10 2 , (1) Observations of the program objects (Table 1) were made in fij λ March 2009, using the high-resolution spectrograph UVES where λ is the wavelength in [A],˚ fij is the absorption os- (UV-Visual Echelle Spectrograph) of the VLT fed by the cillator strength. C2 lines are mostly lying on the linear Kueyen telescope of the ESO Paranal Observatory, Chile. part of the curve of growth. In this work, only a few lines That part of work is a continuation of our previous paper of HD 149757 or HD 154368 are optically thick and curve (Ka´zmierczak et al. 2010) based on archive UVES spectra. of growth method was applied for the derivation of col- The spectral analysis and following calculation of column umn densities. The turbulent velocity was serched through densities, rotational temperatures of C2 and modeling gas the minimalization of the dispersion of column densities for kinetic temperatures and density of the collisional partners each level. We checked various values of the velocity dis- were made in the same way as it was in details described in − persion parameter (b = 0; 0.3; 0.5; 0.7; 1; 1.5kms 1) and (Ka´zmierczak et al. 2010). − 0.5 km s 1 was found to give the lowest dispersion. This All of the selected targets have one C2 Doppler com- value was consequently applied to all of the program stars. ponent at the resolution (R∼85 000) of the observational It is consistent with the value derived in other studies of material. Generally, we cannot exclude the existence of mul- molecular absorptions (e.g. Gredel et al. (1991); Crawford tiple closely spaced components, moreover it is very likely (1997)). that toward all of the stars there are at least two different The energies of the lower rotational level were de- interstellar clouds. The presence of weak components should termined using molecular constants of Marenin & Johnson not contaminate weak C2 lines, so it is not so big problem (1970). The wavelengths are generally determined from to measure dicarbon molecule features, but when we want laboratory wave numbers of Chauville et al. (1977) and to make a comparison with other molecules or DIBs we have Ballik & Ramsay (1963b) converted to air wavelengths us- to be careful. Because of that, not all objects from Table 1 ing Edlen’s formula following Morton (1991). Wavelengths were used to analyse diffuse interstellar bands (e.g. toward of three lines R(2), P(2) and P(4) of the (2,0) band, absent HD 148379 there are more than one dominant Doppler com- in (Chauville et al. 1977), were computed with Douay et al. ponent in CH 4300 A˚ and KI 7698 A˚ interstellar lines). (1988) spectroscopic constants. According to Douay et al. Part of data reduction was made with the DECH20T (1988) the line positions calculated with their constants code (Galazutdinov 1992) and with IRAF1 which was also should be more accurate than the previous measurements.