---
title: 'PEACHES: Protostellar Chemistry in Perseus'
url: https://www.emergentmind.com/topics/peaches-survey
type: topic
---

# PEACHES: Protostellar Chemistry in Perseus

The Perseus ALMA Chemistry Survey (PEACHES) is a cloud-wide, unbiased ALMA Band 6 survey of embedded low-mass protostars in the Perseus molecular cloud. Its purpose is to place the chemistry of Class 0/I sources on a statistical footing by observing a uniform sample with common angular resolution, spectral coverage, and analysis methods. Across the PEACHES series, the survey has been used to quantify the incidence of complex organic molecules (COMs), characterize sulfur-bearing species linked to accretion and ejection, and map hydrocarbon emission associated with photochemically active outflow cavities, thereby establishing Perseus as a benchmark environment for low-mass protostellar chemistry [2101.11009] [2309.05839] [2508.03918].

## 1. Survey definition, region, and source census

Perseus is one of the nearest actively star-forming molecular clouds, spanning about 10 pc on the sky and hosting a large, well-characterized population of embedded protostars in NGC 1333, IC 348, L1448, L1455, B1, and B5. This makes it suitable for a cloud-wide chemistry survey in which source selection is not driven by prior knowledge of molecular richness. The core PEACHES sample comprises 50 embedded Class 0/I protostars selected using the criteria of all Class 0/I sources with $L_{\rm bol} > 1\,L_\odot$ (or $0.7\,L_\odot$ in B1 and B5) and envelope mass $> 1\,M_\odot$, with positions and multiplicity updated from VANDAM. These targets were observed in 37 ALMA fields, yielding 51 continuum peaks; L1448 IRS 2E and SVS 3 are continuum nondetections, while EDJ2009-237, Per-emb 60, and EDJ2009-172 show no lines and were excluded from spectral modeling but retained in detection statistics. For consistent angular-to-linear scaling, the survey adopts a uniform distance of 300 pc, although later PEACHES work notes distance variations across Perseus of 234–331 pc [2101.11009] [2309.05839].

The survey design addressed a limitation of earlier low-mass hot-corino and COM studies, which were dominated by small, heterogeneous samples and targeted follow-up of already known chemically rich sources. In PEACHES, the emphasis is instead on population-level detection fractions, inter-species correlations, and links between chemistry and source structure. Later papers in the series retained this statistical perspective while extending the chemistry from COMs to sulfur-bearing species and C$_2$H.

## 2. Observational architecture and analytical formalism

The original PEACHES observations were carried out in two ALMA Band 6 projects, 2016.1.01501.S and 2017.1.01462.S. Each project used 13 spectral windows: 12 narrow windows with 122 kHz channel spacing, corresponding to about $0.14$–$0.15\ {\rm km\ s^{-1}}$, and one wide spectral window for continuum with 976.6 kHz channels, corresponding to about $1.19\ {\rm km\ s^{-1}}$. Baselines were about 15–919 m in one project and 15–1231 m in the other, producing typical synthesized beams of about $0.6'' \times 0.4''$, or roughly $180 \times 120$ au at 300 pc. Imaging used CASA standard calibration, `tclean` with `robust = 0.5`, a multiscale deconvolver, and primary-beam correction. Continuum images were cleaned to 0.008 Jy and line cubes to 0.022 Jy [2101.11009].

Subsequent PEACHES papers preserved this high-resolution Band 6 framework while adapting the analysis to specific molecular families. The sulfur survey analyzed 50 Class 0/I sources with an average angular resolution of about $0.6''$, total on-source integration of about 10 min per target, and self-calibration of line data from the continuum for most sources. The C$_2$H study combined the PEACHES 12 m data with new ACA observations and used SMA/MASSES CO maps to trace outflows; 12 m-only maps were used for morphology, while 12 m+ACA cubes were used for column density recovery [2309.05839] [2508.03918].

Across the series, column-density analysis relied on standard LTE and non-LTE formalisms. In optically thin LTE,
$$
N_u = \frac{8\pi k \nu^2}{h c^3 A_{ul}} W,
$$
$$
N_{\rm tot} = \frac{N_u\,Q(T_{\rm ex})}{g_u}\,e^{E_u/(k_B T_{\rm ex})},
$$
and the rotational-diagram form is
$$
\ln\!\left(\frac{N_u}{g_u}\right) = \ln\!\left(\frac{N_{\rm tot}}{Q(T_{\rm ex})}\right) - \frac{E_u}{k_B T_{\rm ex}}.
$$
Abundances are written as
$$
X({\rm species}) = \frac{N({\rm species})}{N({\rm H_2})}.
$$
In practice, PEACHES I modeled COMs with `xclass`, which includes optical-depth effects; the sulfur analysis used LTE for SO and $^{34}$SO but non-LTE RADEX grids for SO$_2$; and the C$_2$H analysis fit hyperfine-resolved opacity directly rather than assuming optically thin emission.

## 3. Complex organic molecules and the hot-corino baseline

PEACHES I established the survey’s COM baseline by identifying O-bearing COMs such as CH$_3$OH, CH$_3$OCHO, CH$_3$OCH$_3$, CH$_3$CHO, C$_2$H$_5$OH, and cis-CH$_2$OHCHO, together with N-bearing COMs such as CH$_3$CN, CH$_3$CH$_2$CN, NH$_2$CHO, and CH$_2$DCN. A species identification required at least one unblended line with S/N $> 3$ and agreement with synthetic LTE spectra. On this basis, 29 of 50 sources, or 58%, show COM emission concentrated within $\lesssim 300$ au of the continuum peaks. CH$_3$OH is detected in 28 sources, or 56%; N-bearing COMs in 20 sources, or 40%; and methyl formate in 14 sources, while the abstract summarizes the CH$_3$OCHO incidence as 32% [2101.11009].

The COM results are notable less for a single molecular inventory than for the coherence of the inter-species trends. CH$_3$OH rotational diagrams yield $T_{\rm rot} \sim 140$–350 K, consistent with hot inner-envelope conditions. When column densities are normalized by the continuum brightness temperature $T_b$, which is used as a proxy for gas column density, CH$_3$OH and CH$_3$CN show a tight correlation spanning more than two orders of magnitude, with Pearson $r \approx 0.87$. Among the four most frequently detected COMs—CH$_3$OH, CH$_3$CN, CH$_3$OCHO, and CH$_3$OCH$_3$—pairwise correlations have $r \approx 0.76$–0.93, and the median $r$ across correlations with all COM species is about 0.86. By contrast, COM detectability shows no obvious dependence on averaged continuum brightness temperature, $L_{\rm bol}$, or $T_{\rm bol}$. Ratios of larger COMs such as CH$_3$OCHO and CH$_3$OCH$_3$ to smaller COMs such as CH$_3$OH and CH$_3$CN increase with the inferred gas column density traced by $T_b$, suggesting that denser envelopes favor the production or retention of more complex species [2101.11009].

In later comparison work, PEACHES also became the reference sample for hot-corino incidence. There, hot corinos are defined physically as compact ($\leq 100$ au), hot ($\geq 100$ K), dense ($\geq 10^7\ {\rm cm^{-3}}$) regions enriched in iCOMs, while the operational PEACHES diagnostic is the presence of warm methanol emission. On that basis, the Perseus hot-corino fraction is reported as $(56 \pm 14)\%$ in the main text of the Orion comparison paper and as $\sim 60\%$ in its abstract, making PEACHES the high-incidence benchmark for matched environmental comparisons [2202.13835].

## 4. Sulfur-bearing species as diagnostics of inner envelopes, shocks, and outflows

PEACHES III extended the survey to simple sulfur-bearing species—CS, SO, $^{34}$SO, and SO$_2$—in the same 50-source Perseus sample, consisting of 36 Class 0 and 14 Class I objects. The observations targeted CS $5$–$4$ at 244.9356 GHz, SO $6_6$–$5_5$ at 258.2558 GHz, SO $7_6$–$6_5$ at 261.8437 GHz, $^{34}$SO $5_6$–$4_5$ at 246.6636 GHz, and SO$_2$ $14_{0,14}$–$13_{1,13}$ at 244.2542 GHz, with spectral resolutions of $0.14$–$0.15\ {\rm km\ s^{-1}}$ for CS, SO, and SO$_2$, and $1.2\ {\rm km\ s^{-1}}$ for $^{34}$SO [2309.05839].

| Species | Class 0 | Class I |
|---|---:|---:|
| CS | 97% | 71% |
| SO (either transition) | 86% | 57% |
| $^{34}$SO | 31% | 36% |
| SO$_2$ | 44% | 43% |

The morphologies separate naturally into ejection-tracing and inner-envelope components. CS commonly shows centrally peaked plus extended emission and frequently traces outflow cavities; in some sources it is perpendicular to outflows, consistent with wider cavity walls or infalling-envelope structures. SO traces both compact and extended emission, with clear outflow structures in many Class 0 sources. By contrast, $^{34}$SO and SO$_2$ are predominantly compact around the protostar, although localized extensions occur along outflows or perpendicular structures. CS often shows self-absorption, making its integrated fluxes and column densities lower limits.

The most distinctive sulfur result is the behavior of the SO/$^{34}$SO ratio. Under optically thin conditions one expects $N({\rm SO})/N(^{34}{\rm SO}) \approx {}^{32}{\rm S}/{}^{34}{\rm S} \approx 22$, but the observed column-density ratios are systematically lower than 22, implying significant optical depth in SO. The lowest values, $\leq 1.5$, occur in the COM-rich sources Per-emb 13, 12B, 27, 29, and 44. The ratio is therefore proposed as a tracer of dense inner-envelope gas with high SO opacity and as a proxy for conditions conducive to COM emission. SO$_2$ is likewise associated with warm compact gas: typical $N({\rm SO_2}) \lesssim 10^{15}\ {\rm cm^{-2}}$ under optically thin assumptions, but four sources show lower limits $\geq 2\times10^{15}\ {\rm cm^{-2}}$, indicating optically thick SO$_2$. Where both CH$_3$OH and SO$_2$ are detected, $X({\rm CH_3OH})$ is 10–100 times $X({\rm SO_2})$, and the two abundances are positively correlated, pointing to co-spatial warm-gas chemistry [2309.05839].

The sulfur survey also tied chemistry to source dynamics and environment. Sources with SO$_2$ detections and CS/SO aligned with collimated outflows are usually COM-rich, whereas sources with SO$_2$ but CS/SO perpendicular to outflows generally lack COMs. SO$_2$ abundances in the Perseus sample are, on average, two orders of magnitude lower than in Ophiuchus and broadly comparable to Taurus, suggesting that gas-phase sulfur depletion depends on the external UV field. In evolutionary terms, SO$_2$ in Class 0 sources traces a high column density of warm material near the protostar, while in more evolved systems it can trace shocks or UV-irradiated material.

## 5. C$_2$H and photochemically active cavity walls

PEACHES IV shifted the survey emphasis from iCOM-rich hot gas and sulfur-bearing shock tracers to hydrocarbon chemistry. Using the PEACHES 12 m data together with new ACA observations, the study examined C$_2$H toward an unbiased sample of 35 Class 0/I low-mass protostars in Perseus. The targeted transition was the C$_2$H $N=3$–2, $J=5/2$–$3/2$ hyperfine group near 262.06 GHz, observed together with SO at 261.8437 GHz, with the ACA spectral window sampled at 61.035 kHz, or about $0.07\ {\rm km\ s^{-1}}$ [2508.03918].

The principal result is morphological rather than merely statistical. C$_2$H is detected in 30 of 35 protostars, and SO is also detected in 30 of 35. Many sources show extended C$_2$H emission over a few arcseconds in narrow, conical, or “X-wing” structures aligned with CO outflow cavity walls and bases. Intensity profiles show that C$_2$H generally peaks off the continuum and exhibits dips or minima at the continuum peak. In 12 of the 14 sources with clearly discernible bipolar CO outflows, average intensity profiles reveal a robust anti-correlation between C$_2$H and SO: C$_2$H avoids SO-bright regions and instead peaks where SO is weak, and vice versa. The physical interpretation is that C$_2$H traces photochemically active, oxygen-poor gas in FUV-irradiated cavity walls, whereas SO traces shocked, oxygen-rich gas. Geometry modulates how clearly this pattern appears in individual sources, but the sample-level trend is systematic [2508.03918].

Column densities were derived from pixel-by-pixel LTE fits to the hyperfine structure. Each hyperfine component was modeled as a Gaussian with a common rotational temperature $T_r$, common velocity dispersion, and common $V_{\rm lsr}$. Because the hyperfine set shares the same upper-state energy, $T_r$ cannot be constrained directly and was fixed at 20 K; the inferred $N({\rm C_2H})$ changes by less than 5% if $T_r$ is varied within 10–50 K. The total opacity profile is modeled as the sum over hyperfine components, and the emergent intensity is
$$
I_\nu = [B_\nu(T_r)-B_\nu(T_{\rm bg})]\,[1-e^{-\tau_\nu}]\,\Omega,
$$
with $T_{\rm bg}=2.73$ K. The resulting C$_2$H columns span about $10^{14}$–$10^{15}\ {\rm cm^{-2}}$, with an average across all sources of about $10^{14.3\pm0.2}\ {\rm cm^{-2}}$. No clear correlations were found between the spatial extent of C$_2$H emission and $T_{\rm bol}$, $L_{\rm bol}$, or the deconvolved mm-continuum radius. The implication is that C$_2$H remains extended throughout the embedded phase and that the transition to the compact disk-surface C$_2$H seen in Class II systems must occur rapidly, just after the embedded stage [2508.03918].

## 6. Environmental comparison, caveats, and broader significance

PEACHES acquired additional significance through direct comparison with the ORion ALMA New GEneration Survey (ORANGES). The Orion survey was designed to be analogous to PEACHES, with matched sensitivity, comparable spatial resolution, and comparable spectral setup; the quoted sensitivities are about 22 mJy beam$^{-1}$ for PEACHES and about 24 mJy beam$^{-1}$ for ORANGES. Using warm CH$_3$OH as the practical hot-corino tracer, PEACHES provides a Perseus reference fraction of about 0.56, or reported $(56 \pm 14)\%$, from 50 protostars, whereas ORANGES finds 5 bona fide hot corinos in 19 sources, or reported $(26 \pm 23)\%$, in the Orion OMC-2/3 filament. Perseus is characterized as a low-mass star-forming region that is less illuminated and more shielded from strong external UV radiation, while OMC-2/3 is a massive filament bounded by three H II regions and highly illuminated by UV photons from massive stars. This difference suggests that environment is likely playing a role in shaping protostellar chemistry, although the Orion study emphasizes that improved statistics are required to consolidate the result [2202.13835].

The survey series also clarifies what PEACHES does and does not show. It does not support a simple picture in which chemical richness is set only by luminosity, evolutionary class, or continuum brightness: COM detectability depends neither on averaged continuum brightness temperature nor on $L_{\rm bol}$ and $T_{\rm bol}$, and C$_2$H morphology likewise shows no clear trend with $T_{\rm bol}$, $L_{\rm bol}$, or continuum radius. At the same time, the PEACHES results are not free of observational or modeling caveats. Dust opacity can obscure submillimeter lines in bright continuum sources; CS and SO are often optically thick; compact $^{34}$SO, SO$_2$, and hot-corino emission can be beam-diluted; and single-transition or single-$J$ analyses require assumptions about excitation that can only be relaxed with additional lines or higher angular resolution [2101.11009] [2309.05839] [2508.03918].

Taken together, the PEACHES papers establish a chemically differentiated picture of embedded low-mass protostars in Perseus. COMs are common but not universal, and their inter-species correlations are strong even when their absolute abundances vary by orders of magnitude. Sulfur-bearing species separate compact warm inner regions from shocked outflow structures and reveal an environmental dependence of sulfur depletion. C$_2$H identifies FUV-processed, oxygen-poor cavity-wall gas that is spatially segregated from SO-rich shocked layers. The survey therefore functions both as a statistical atlas of Perseus protostellar chemistry and as a reference frame for testing how local environment, UV irradiation, outflow geometry, and warm-column density shape the molecular initial conditions of disk and comet formation.

Source: https://www.emergentmind.com/topics/peaches-survey