G+0.633: Shock-Processed CMZ Cloud

• G+0.633 is a molecular cloud in the CMZ characterized by warm gas, moderate density, and shock-dominated, chemically enriched conditions without prominent high-mass young stellar objects.
• Broad spectral surveys using Yebes, IRAM, and APEX reveal three distinct velocity components and quantify temperature, density, and molecular abundances in detail.
• The detection of benzonitrile and aromatic molecules under shock conditions illustrates unique gas-phase chemical pathways, informing our understanding of early cluster formation.

G+0.633-0.0604, usually abbreviated G+0.633, is a molecular cloud in the Central Molecular Zone (CMZ) of the Milky Way, located in the southern part of the Sgr B2 complex and identified as a shock-dominated, chemically rich, largely non-hot-core environment (Andrés et al., 1 Jul 2026). It has become a reference source for Galactic-centre astrochemistry because it combines warm gas, moderate density, strong evidence for large-scale shocks, and an absence of clear high-mass young stellar object signatures at its main velocity component, while also hosting aromatic chemistry exemplified by the detection of benzonitrile, c-C$_6$H$_5$CN (Rivilla et al., 27 Apr 2026). In current CMZ studies, G+0.633 is commonly treated as a southern counterpart to G+0.693-0.027 and as a laboratory for testing how shocks, turbulence, cosmic-ray processing, and molecular complexity interact before or at the onset of cluster formation.

1. Galactic setting and source identification

G+0.633 lies in the “Deep South” envelope of Sgr B2, denoted Sgr B2(DS), at the southern edge of the complex. The single-dish pointing coordinates used in the deep spectral surveys are

$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$

Within the north-south Sgr B2 structure, it is located about $2'$ south of Sgr B2(S), while G+0.693 lies to the north of the main Sgr B2 ridge (Andrés et al., 1 Jul 2026). A separate measurement used in the benzonitrile study places G+0.633 about $250''$ from G+0.693, corresponding to a projected $\sim 10.4$ pc at 8.2 kpc (Rivilla et al., 27 Apr 2026).

The source was selected because it coincides with the brightest $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$ peak in the southern Sgr B2 region. That positional coincidence is significant because HNCO is treated in these studies as a tracer of low-velocity shocks. G+0.633 is therefore not merely another dense CMZ sightline, but a deliberately chosen target at a chemically and dynamically active location within the Sgr B2 environment (Andrés et al., 1 Jul 2026).

Its environmental characterization is internally consistent across the two cited studies. G+0.633 is described as warm, moderately dense, chemically rich, and extended on single-dish beam scales. The gas is warm while the dust is cold, with CMZ dust temperatures around $\sim 20$ K, implying that the gas heating is not dominated by embedded massive star formation but instead by shocks, turbulence, and enhanced cosmic-ray ionization (Rivilla et al., 27 Apr 2026). This places G+0.633 in the broader class of quiescent but harsh CMZ clouds: non-hot-core, shock-processed, and chemically rich.

2. Observational basis

The cloud has been characterized with an ultra-broadband set of single-dish spectral surveys plus wide-field mapping. The physical-property study combines Yebes 40 m, IRAM 30 m, and APEX 12 m observations covering $\sim 100$ GHz across the 31–275 GHz range (Andrés et al., 1 Jul 2026). The benzonitrile analysis uses the Yebes 40 m ultra-deep Q-band survey from 31.07 to 50.42 GHz and complementary IRAM 30 m 3 mm data for related nitrile chemistry (Rivilla et al., 27 Apr 2026).

The Yebes 40 m Q-band observations were carried out in position-switching mode, with about 90 hours on source over 25 nights from February to April 2024. The half-power beam width ranges from $\sim 55''$ at 31 GHz to $_5$0 at 50 GHz, and the spectra were smoothed to 256 kHz channels, corresponding to $_5$1–$_5$2 across the band. The rms noise reaches $_5$3–$_5$4 mK in $_5$5, and intensities are reported in antenna temperature because the emission is treated as extended over the primary beam (Rivilla et al., 27 Apr 2026).

The broader physical characterization additionally uses IRAM 30 m 3 mm spectroscopy over 72–116.6 GHz and APEX 12 m coverage in the 210–275 GHz domain, together yielding a highly sensitive line survey suitable for multitransition thermometry and density diagnostics. Large-scale $_5$6 IRAM 30 m mosaics of Sgr B2 in $_5$7, $_5$8, and $_5$9 place G+0.633 in its environmental context and show how its emission connects to the rest of the Sgr B2 complex (Andrés et al., 1 Jul 2026).

These observing strategies matter because G+0.633 is chemically rich but kinematically structured. The combination of deep pointed surveys and mosaics permits both line-by-line modeling at the source position and spatial discrimination between the cloud’s internal components and neighboring Sgr B2 structures.

3. Kinematics and gas physical conditions

Spectra toward G+0.633 show three distinct velocity components along the line of sight. The main one, C1, defines the cloud morphologically; C2 is a broader, fainter component linked to the same larger-scale interaction; and C3 is a high-velocity component regarded as kinematically unlinked to the main G+0.633 structure and instead associated with large-scale CMZ dynamics (Andrés et al., 1 Jul 2026).

Component	Kinematics	Gas properties
C1	$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$0 km s$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$1, FWHM $$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$2 km s$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$3	$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$4–90 K, $$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$5, $$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$6–$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$7
C2	$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$8 km s$$\alpha_{\rm ICRS} = 17^{\rm h}47^{\rm m}21.18^{\rm s},\qquad \delta_{\rm ICRS} = -28^\circ25'36.99''.$$9, FWHM $2'$0 km s$2'$1	$2'$2–90 K, $2'$3, $2'$4–$2'$5
C3	$2'$6 km s$2'$7, FWHM $2'$8 km s$2'$9	$250''$0–90 K, $250''$1, $250''$2–$250''$3

The $250''$4 column densities were derived from optically thin CO isotopologues using LTE modeling and the Galactic Centre isotopic ratios

$250''$5

together with $250''$6, which gives

$250''$7

The resulting CO columns are $250''$8 for C1, $250''$9 for C2, and $\sim 10.4$0 for C3 (Andrés et al., 1 Jul 2026).

The kinetic temperature was inferred from $\sim 10.4$1 and $\sim 10.4$2 $\sim 10.4$3-ladder analyses. The rotational-diagram relation used is

$\sim 10.4$4

with $\sim 10.4$5 serving as a proxy for $\sim 10.4$6 in these symmetric tops. The volume density was constrained through non-LTE RADEX modeling of $\sim 10.4$7 transitions from $\sim 10.4$8–3 to 12–11. The best-fit density ranges are modest by dense-core standards, around $\sim 10.4$9, and this subcritical regime explains why many molecules show low excitation temperatures despite substantially warmer kinetic temperatures (Andrés et al., 1 Jul 2026).

In physical terms, G+0.633 is therefore a warm, subthermally excited CMZ cloud with extended molecular emission and only intermediate density. That combination is characteristic of shock-heated CMZ gas rather than compact hot-core gas.

4. Shock-dominated chemistry and environmental interpretation

The source is identified as shock-dominated on both chemical and morphological grounds. HNCO is central to that diagnosis. In G+0.633, the main component C1 has $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$0 and $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$1, while C2 and C3 have lower but still measurable abundances of $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$2 and $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$3, respectively (Andrés et al., 1 Jul 2026). The strength of HNCO, together with the detection of bright SiO and other shock tracers in the broader G+0.633/G+0.693 context, is taken as evidence for large-scale shocks.

The spatial information reinforces that interpretation. In the IRAM 30 m mosaics, the C1 emission in $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$4, $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$5, and $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$6 peaks at G+0.633 itself. HNCO and ethanol trace lobes around the Sgr B2 low-velocity “hole,” while C2 extends northward toward G+0.693 and the Sgr B2 cores. C1 therefore delineates G+0.633 as a physically distinct cloud, whereas C2 functions as a spatial bridge toward the northern interaction zone (Andrés et al., 1 Jul 2026).

The kinematic interpretation places G+0.633 within the cloud-cloud collision believed to shape Sgr B2. In this picture, C1 traces the disturbed low-velocity component at $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$7–50 km s$\mathrm{HNCO}\,4_{0,4}-3_{0,3}$8, C2 traces the higher-velocity side of the interaction near $\mathrm{HNCO}\,4_{0,4}-3_{0,3}$9–80 km s$\sim 20$0, and C3 belongs instead to a separate high-velocity CMZ stream. The narrow C1 linewidth of about 10 km s$\sim 20$1 is notable because it is smaller than the broader linewidths often reported for G+0.693 and other CMZ clouds. This has been interpreted as possible turbulence dissipation in the component most likely to evolve toward collapse (Andrés et al., 1 Jul 2026).

The source also shows Class I $\sim 20$2 masers, especially a 36 GHz line much stronger than the 44 GHz line, a pattern associated with shock-dominated and non-star-forming environments. By contrast, there is no evidence at G+0.633 for bright ultracompact H II regions, strong free-free compact sources, Class II $\sim 20$3 masers, OH masers, $\sim 20$4 masers, or embedded hot cores at the position of C1 (Andrés et al., 1 Jul 2026). This combination is important because it distinguishes shock processing from classical embedded massive-star heating.

5. Aromatic chemistry and benzonitrile

G+0.633 is one of two CMZ clouds in which benzonitrile, c-C$\sim 20$5H$\sim 20$6CN, was detected in the 31–50 GHz Yebes 40 m survey, establishing the presence of aromatic ring chemistry in a warm Galactic-centre molecular cloud (Rivilla et al., 27 Apr 2026). The detection is based on multiple Q-band transitions, with the stacked analysis using 9 of the cleanest lines after subtraction of contaminating species. The stacked profile reaches a per-channel $\sim 20$7 peak of about 8 and an integrated signal-to-noise ratio

$\sim 20$8

which the authors treat as statistically robust.

The LTE analysis was performed with MADCUBA-SLIM using the CDMS 103501 entry. Because free fitting produced convergence problems in G+0.633, the benzonitrile fit fixed $\sim 20$9 K and $\sim 100$0, leaving only $\sim 100$1 and $\sim 100$2 free. The resulting parameters are

$\sim 100$3

$\sim 100$4

With $\sim 100$5, the abundance is

$\sim 100$6

The non-LTE analysis used RADEX with benzonitrile–He collisional rates extended up to $\sim 100$7 and $\sim 100$8 K, scaled by 1.4 to approximate $\sim 100$9. For G+0.633 the best match is obtained for

$\sim 55''$0

confirming that the excitation is subthermal, $\sim 55''$1, while still allowing an effective single-temperature LTE description with $\sim 55''$2 K (Rivilla et al., 27 Apr 2026).

The aromatic detection is especially significant when compared to linear cyanopolyynes. In the C1 component, aligned in velocity with benzonitrile, the derived $\sim 55''$3 column density is $\sim 55''$4, corresponding to $\sim 55''$5. The resulting abundance ratio is

$\sim 55''$6

That ratio is lower than the 4.5–30 range compiled for colder Galactic clouds. The authors therefore argue that simple direct chemical linkage between $\sim 55''$7 and benzonitrile is unlikely, and that the CMZ environment favors aromatic chemistry relative to long carbon chains (Rivilla et al., 27 Apr 2026).

For formation, the preferred mechanism is the gas-phase neutral-neutral reaction

$\sim 55''$8

described as exothermic and barrierless, with laboratory kinetics showing a rate nearly independent of temperature between 10 and 100 K. Alternative bottom-up precursor schemes involving the $\sim 55''$9- and $_5$00-isomers of $_5$01 are disfavored observationally because those species were not detected and the upper limits place them at factors $_5$02–4.5 below benzonitrile in G+0.633 (Rivilla et al., 27 Apr 2026).

The more difficult question is the origin of benzene itself. Because benzonitrile in G+0.633 has an abundance almost identical to that in cold dark clouds, the paper argues that bottom-up models underpredict the relevant aromatic precursor abundances and that a top-down scenario is more consistent with the observations. In that scenario, shocks, turbulence, and enhanced cosmic rays fragment larger carbonaceous grains or PAHs, generating smaller aromatic units that can then react with CN. This suggests that aromatic chemistry in G+0.633 is not an anomalous by-product of a single pathway, but part of a broader CMZ carbon-processing cycle.

6. Evolutionary status and broader significance

G+0.633 is frequently compared with G+0.693 and described as an “astrochemical twin,” but one that may be at an earlier dynamical stage (Andrés et al., 1 Jul 2026). The resemblance lies in the warm gas, modest density, chemical richness, shock signatures, and apparent absence of strong ongoing massive star formation at the main component. The difference is that G+0.633-C1 has a narrower linewidth and appears more spatially confined, making it a candidate for a very early protocluster phase.

That interpretation is tied to column density as well. In Sgr B2, high-mass star formation is observed where $_5$03, and no young stellar objects are seen below $_5$04. The C1 value in G+0.633 is $_5$05, about a factor of 2–3 below that level (Andrés et al., 1 Jul 2026). The inference advanced in the physical-property study is therefore cautious: C1 may represent gas in which shocks have heated and enriched the medium and turbulence has partly dissipated, but large-scale gravitational collapse into an active massive protocluster has not yet been demonstrated.

The benzonitrile study broadens the significance of the source beyond CMZ dynamics. G+0.633 shows that aromatic rings can survive and remain abundant in a cloud subject to high kinetic temperatures, shocks, and enhanced cosmic-ray ionization (Rivilla et al., 27 Apr 2026). The direct gas-phase carbon contribution of benzonitrile itself is small: $_5$06 or about $_5$07 of all carbon atoms if the total interstellar carbon abundance is $_5$08 per H nucleus. However, the same study argues that the detection is chemically diagnostic rather than carbon-budget dominant: if larger aromatic species are present at comparable relative levels, their cumulative contribution could become non-negligible.

Taken together, the current view of G+0.633 is that of a shock-processed CMZ cloud in which chemical enrichment, aromatic stability, and pre-cluster dynamical evolution can be studied simultaneously. It is important not because it exemplifies only one phenomenon, but because it connects several: cloud-cloud collision kinematics, low-velocity shock chemistry, subthermal excitation in warm gas, the persistence of aromatic molecules under harsh CMZ conditions, and the possibility that parts of Sgr B2(DS) mark the next episode of cluster formation in the complex (Andrés et al., 1 Jul 2026).