2025ApJ...987..209G

Analysis of very high signal-to-noise ratio (S/N) spectra of reddened stars allowed us to determine an upper limit (UL) of the column density N(C₄). We developed a robust calibration of the N(C₄) based on the C₄ gas-phase spectrum profile at 3789 Å. The UL of the column density can be calculated using the equation N_UL(C₄) = 8.963 × 10¹⁵/(S/N), where S/N is in the wavelength range ∼3790 Å. Nondetection of this molecule was confirmed in stars exhibiting a high abundance of C₂ and C₃ molecules. According to our calibration, assuming an optimistic estimate of the ratio $\frac{N({{\rm{C}}}_{3})}{N({{\rm{C}}}_{4})}$ = 10, even for the star HD 147889 with the highest known to date column density of C₃, the S/N required for detecting the C₄ molecule is about 12,000 per pixel, which is a challenging task. Despite this tough conclusion regarding the possibility of detecting the interstellar molecule C₄, we also tried to find a link between known lines of C₅ at 4975 and 5109 Å and weak interstellar features. The expected C₅ 4975 Å feature coincides well with weak diffuse interstellar band (DIB) 4975, while the search for the 5109 Å C₅ band is badly conditioned by the overlap with a broad spectral feature, which is a blend of stellar and interstellar lines. However, we found a rather good correlation between the equivalent width of DIB 4975 and N(C₂), making the search for C₅ not hopeless. If DIB 4975 is formed by C₅, its column density is only 2 times lower than that of C₃.