[Paper Review] ISM Properties in Low-Metallicity Environments III. The Dust Spectral Energy Distributions of II Zw 40, He 2-10 and NGC 1140

F. Galliano, S. C. Madden+3 more|ArXiv.org|Jan 28, 2005

Astrophysics and Star Formation StudiesPhysics and Astronomy71 references123 citations

TL;DR

This study models the dust spectral energy distributions (SEDs) of three low-metallicity dwarf galaxies—II Zw 40, He 2-10, and NGC 1140—using multiwavelength submillimeter to infrared data. It reveals that these galaxies host extremely small dust grains (~3–4 nm), a severe lack of polycyclic aromatic hydrocarbons (PAHs), and a significant population of very cold dust (5–9 K), accounting for 40–80 % of the total dust mass, which explains the observed millimeter SED excess. The findings challenge Galactic dust assumptions and highlight the role of shocks and hard radiation in shaping dust properties in low-metallicity environments.

ABSTRACT

We present new 450 and 850 micron SCUBA data and 1.3 mm MAMBO data of the dwarf galaxies II Zw 40, He 2-10 and NGC 1140. Additional ISOCAM, IRAS as well as ground based data are used to construct the observed mid-infrared to millimeter spectral energy distribution of these galaxies. These spectral energy distributions are modeled in a self-consistent way, as was achieved with NGC 1569 (Galliano et al., 2003), synthesizing both the global stellar radiation field and the dust emission, with further constraints provided by the photoionisation of the gas. Our study shows that low-metallicity galaxies have very different dust properties compared to the Galaxy. Our main results are: (i) a paucity of PAHs which are likely destroyed by the hard penetrating radiation field, (ii) a very small (3-4 nm) average size of grains, consistent with the fragmentation and erosion of dust particles by the numerous shocks, (iii) a significant millimetre excess in the dust spectral energy distribution which can be explained by the presence of ubiquitous very cold dust (T=5-9 K) accounting for 40 to 80 % of the total dust mass, probably distributed in small clumps. We derive a range of gas-to-dust mass ratios between 300 and 2000, larger than the Galactic values and dust-to-metals ratios of 1/30 to 1/2. The modeled dust size distributions are used to synthesize an extinction curve for each galaxy. The UV slopes of the extinction curves resemble that observed in some regions in the Large Magellanic Cloud. The 2175 angstrom bumps of the modeled extinction curves are weaker than that of the Galaxy, except in the case of II Zw 40, where we are unable to accurately constrain the 2175 angstrom bump carrier.

Motivation & Objective

• To understand how dust properties in low-metallicity dwarf galaxies differ from those in the Milky Way.
• To model the full dust SED from mid-infrared to submillimeter wavelengths in a self-consistent way, including both emission and extinction.
• To investigate the physical mechanisms shaping dust grain size distributions and emission characteristics in extreme environments with low metallicity and strong radiation fields.
• To determine the contribution of very cold dust (5–9 K) to the total dust mass and its implications for the observed millimeter SED excess.
• To derive gas-to-dust mass ratios and dust-to-metal ratios, and to assess the validity of Galactic dust models in low-metallicity environments.

Proposed method

• Combined multiwavelength data from SCUBA (450 and 850 µm), MAMBO (1.3 mm), ISOCAM, IRAS, and ground-based observations to construct complete dust SEDs.
• Used a self-consistent modeling approach that simultaneously synthesizes the global stellar radiation field and dust emission, with constraints from photoionization of the gas.
• Employed the Désert et al. (1990) dust model with variable grain size distributions and radiation field parameters to fit the observed SEDs.
• Explored the impact of temperature-dependent absorption efficiency $ Q_{\text{abs}}(\lambda, a, T) $ for silicate grains based on Agladze et al. (1996) measurements to test for anomalous emissivity in cold grains.
• Constrained the dust size distribution by fitting the observed SEDs while varying both the radiation field and grain size parameters, avoiding degeneracy between solutions.
• Derived extinction curves from the modeled dust size distributions and compared their UV slopes and 2175 Å bumps to Galactic and LMC-like extinction laws.

Experimental results

Research questions

• RQ1 How do dust grain size distributions in low-metallicity dwarf galaxies differ from those in the Milky Way?
• RQ2 What causes the observed millimeter SED excess in II Zw 40, He 2-10, and NGC 1140, and can it be explained by cold dust components?
• RQ3 Why are PAH features suppressed in these galaxies, and what role does the hard radiation field play?
• RQ4 To what extent do the observed dust properties challenge the assumption of Galactic dust characteristics in extragalactic SED modeling?
• RQ5 How do the derived gas-to-dust mass ratios and dust-to-metal ratios compare to Galactic values in low-metallicity environments?

Key findings

• The dust grain size distribution in these low-metallicity galaxies is dominated by extremely small grains with an average size of ~3–4 nm, consistent with fragmentation and erosion by shocks.
• Polycyclic aromatic hydrocarbons (PAHs) are severely depleted or absent, likely due to destruction by the hard, penetrating radiation field.
• A significant millimeter SED excess is explained by the presence of very cold dust at 5–9 K, which accounts for 40–80 % of the total dust mass and is likely distributed in small, cold clumps.
• The gas-to-dust mass ratio ranges from 300 to 2000, significantly higher than the Galactic value, and the dust-to-metal ratio is between 1/30 and 1/2, indicating inefficient dust production or growth.
• The modeled extinction curves show UV slopes similar to those observed in the Large Magellanic Cloud, with weakened 2175 Å bumps compared to the Galaxy, except in II Zw 40 where the bump is unconstrained.
• Temperature-dependent optical properties of silicate grains (e.g., $ Q_{\text{abs}} $) do not significantly alter the SED fit and cannot explain the submillimeter excess, suggesting other physical mechanisms or dust compositions may be at play.