Backlighting young stellar objects in the Central Molecular Zone: an ensemble-averaged abundance structure of methanol ices

TL;DR

This work tackles the challenge of characterizing ices in the Central Molecular Zone (CMZ) by exploiting backlighting from foreground giants to derive an ensemble-averaged methanol-ice abundance profile in YSO envelopes. By combining new $L$-band Gemini/GNIRS spectra of 15 red CMZ sources with $K$-band and Spitzer/IRS mid-IR data, the authors model the SEDs to constrain foreground extinction ($A_K^{*}$, $A_K^{MIR}$) and solid $H_2O$ columns, and they directly measure solid $CH_3OH$ through the $3.535\,\mu$m feature. They find CH$_3OH$ in CH$_3OH$-CO$_2$ ice mixtures at $2$%--$5$% (mean $3.3\pm0.7\%$), lower than typical Galactic-disk values, with the CH$_3$OH/CO$_2$ shoulder ratio rising from $\sim10\%$ in inner envelopes to $\sim30\%$ in outer regions, and a CH$_3$OH/H$_2O$ ratio that declines toward the center. These trends can reflect either intrinsic CMZ chemical differences or, plausibly, heating-driven methanol sublimation in the inner envelopes of a sample biased toward massive YSOs. The ensemble-averaged radial abundance structure revealed here demonstrates the power of backlit spectroscopy to probe ice chemistry at the CMZ's distance and sets the stage for higher-resolution follow-ups (e.g., JWST) to disentangle chemical evolution from thermal processing in extreme Galactic environments.

Abstract

The Central Molecular Zone (CMZ) contains a substantial reservoir of dense molecular gas, where numerous young stellar objects (YSOs) and dense cloud cores have been identified. However, the large distance and severe foreground extinction complicate interpretation of infrared ice absorption features tracing chemical and evolutionary properties of these embedded objects. To better characterise YSOs and dense cores in this region, we combined spectra from multiple YSOs, each likely backlit by a giant star, allowing us to probe their outer layers and derive an ensemble-averaged ice abundance profile. We obtained L-band spectra of 15 point-like sources with extremely red colours using Gemini/GNIRS, enabling measurements of the CH3OH absorption feature at 3.535 micron. To better constrain the foreground extinction and H$_2$O ice column densities, we combined these data with K-band and mid-infrared spectra using NASA/IRTF and Spitzer/IRS. We found that the CH$_3$OH abundance in the CH$_3$OH-CO$_2$ ice mixture is 2 to 5 percent, confirming that it is systematically lower than those typically observed in the Galactic disk. Furthermore, by using the local excess of foreground extinction as a proxy for the projected distance between a backlit source and the centre of a YSO, we found that the CH$_3$OH abundance relative to solid CO$_2$ remains near 10 percent in the inner regions of the envelope, but increases sharply to about 30 percent in the outer regions. The relatively low methanol ice abundance may reflect the unique chemical environment of the CMZ. However, our results offer an alternative interpretation: since our sample is biased towards massive and luminous YSOs, intense heating from the central protostar may have caused substantial sublimation of methanol ice in the inner regions of their envelopes, thereby systematically lowering the observed CH$_3$OH/H$_2$O ice ratios.

Figures (15)

• Figure 1: GNIRS $L$-band spectra of 12 red sources in the CMZ. The observed spectra (red) are vertically offset for clarity and approximately ordered by increasing strength of the broad H$_2$O ice absorption band centred at $3\ \mu$m. To illustrate the varying depth of this feature, the best-fit IRTF/SpeX M5.5III template spectrum rayner:09 is shown in green for each source. These templates were reddened using the foreground extinction estimate derived from our spectral energy distribution (SED) fitting procedure (see Sect. \ref{['sec:sed']}). Together with the $3\ \mu$m ice absorption profile, they reproduce the overall continuum shape and slope of the observed spectra.
• Figure 2: Same as Figure \ref{['fig:gnirs_l']}, but those containing both $K$- and $L$-band spectra from GNIRS observations.
• Figure 3: Composite IR spectrum of a red CMZ object (SSTGC 799887) and modelling of the SED. The observed spectra are shown as a solid red line, constructed by combining near-IR data from Gemini/GNIRS with mid-IR data from Spitzer/IRS. Circles represent photometric measurements from UKIDSS and Spitzer/IRAC, which were used to set the flux scale of the spectra (see text). The blue solid line shows the best-fitting model to the observed spectra, which included the following components: (1) a template of an M5.5-type giant from the IRTF/SpeX spectral library rayner:09 (green line); (2) a two-component blackbody continuum (dashed blue and orange lines); (3) the $3\ \mu$m band of solid H$_2$O ice, using the template from GC IRS 7 jang:22; and (4) the $10\ \mu$m and $18\ \mu$m silicate bands from the GC IRS 3 spectrum kemper:04. Other weaker absorption features were not included in the modelling. Spectra and their best-fitting models for the additional objects are shown in Appendix \ref{['sec:sed2']}.
• Figure 4: Distribution of extinction and H$_2$O ice column densities derived from the global fits to the composite spectra. The two extinction estimates, $A_K^{\mathrm{MIR}}$ and $A_K^{\mathrm{*}}$, come from our composite spectra but rely on different wavelength regimes: the mid-IR spectral shape for $A_K^{\mathrm{MIR}}$ and near-IR template matching for $A_K^{\mathrm{*}}$. Objects with large solid CH$_3$OH column densities ($N > 3 \times 10^{17}\, \mathrm{cm}^{-2}$; see below) are marked with blue open circles. The dashed line in the top panel indicates unity. For comparison, the solid lines in the middle and bottom panels show the empirical relation derived for the Taurus molecular cloud by whittet:01, shifted horizontally by $A_K = 2.2$ mag to account for the average foreground extinction towards the CMZ.
• Figure 5: Schematic illustration of the light-ray geometry probed by the composite spectra. $A_K^{\mathrm{*}}$ reflects extinction constrained mainly by the near-IR, while $A_K^{\mathrm{MIR}}$ is determined primarily from the mid-IR. The quantity $A_K^{\mathrm{fg}}$ represents the foreground extinction estimated from nearby source-free regions, which accounts for dust located along the line of sight in the Galactic disk and within the CMZ. The envelope-only extinction, $A_{K,0}^{\mathrm{*}}$, is defined as $A_K^{\mathrm{*}} - A_K^{\mathrm{fg}}$, isolating the extinction produced by material within the protostellar envelope along the sightline to the background star.