Chemical differentiation in regions of high-mass star formation

Our goals are to compare the CS, N2H and dust distributions in a representative sample of high-mass star forming dense cores and to determine the physical and chemical properties of these cores. We compare the results of CS(5–4) and 1.2 mm continuum mapping of twelve dense cores from the southern hemisphere presented in this work, in combination with our previous N2H(1–0) and CS(2–1) data. We use numerical modeling of molecular excitation to estimate physical parameters of the cores. Most of the maps have several emission peaks (clumps). Mean sizes of 17 clumps, having counterparts in continuum and CS, are 0.30(0.06) pc (continuum) and 0.51(0.07) pc (CS). For the clumps with IRAS sources we derived dust temperatures: 24–35 K, masses: 90–6900 M⊙, molecular hydrogen column densities: (0.7−12.0)×1023 cm−2 and luminosities: (0.6 − 46.0) × 104 L⊙. LVG densities towards CS peaks within the 50 beam (0.56 pc at 2.3 kpc, the average distance of our sample source) vary from source to source in the range: (3–40)×105 cm−3. Masses calculated from LVG densities are higher than CS virial masses and masses derived from continuum data, implying small-scale clumpiness of the cores. The molecular abundances towards IRAS sources in eight objects are X(CS)=(0.3 − 2.7) × 10−9 and X(N2H)=(0.3−4.4)×10. The CS and continuum maps have been compared with each other and with the N2H(1–0) maps. For most of the objects, the CS and continuum peaks are close to the IRAS point source positions. The CS(5–4) intensities correlate with continuum fluxes per beam in all cases, but only in five cases with the N2H(1–0) intensities. The study of spatial variations of molecular integrated intensity ratios to continuum fluxes per beam reveals that I(N2H)/F1.2 ratios drop towards the CS peaks for most of the sources, which can be due to a N2H abundance decrease. For CS(5–4), the I(CS)/F1.2 ratios show no clear trends with distance from the CS peaks, while for CS(2–1) such ratios drop towards these peaks. Possible explanations of these results are considered. The analysis of normalized velocity differences between CS and N2H lines has not revealed indications of systematic motions towards CS peaks.