Background

Most stars in our galaxy don't form in isolation. Instead, they're born within dense stellar clusters in massive star-forming regions, where young protoplanetary disks can be exposed to ultraviolet radiation fields thousands of times stronger than the typical interstellar background. This intense external irradiation fundamentally shapes how these disks evolve, driving powerful photoevaporative winds that strip away material and sculpt the disk structure.

For decades, astronomers have studied the physical effects of this process through observations of iconic systems like the Orion Nebula proplyds and through sophisticated hydrodynamical simulations. We know that externally irradiated disks tend to be more compact, less massive, and shorter-lived than their isolated counterparts. However, while we've made great progress understanding the physical evolution of these disks, their chemical evolution has remained comparatively unexplored. Since most planets likely form in these clustered environments rather than in isolation, understanding how external irradiation affects disk chemistry is crucial for understanding planet formation across the galaxy.

The Challenge

One of the main obstacles has been computational. Chemical models require detailed knowledge of the disk's density structure, temperature distribution, and radiation field. For externally irradiated disks, this means accounting for the photoevaporative wind, which is the outflowing material driven off the disk surface by intense UV radiation. But computing these wind structures requires expensive radiation-hydrodynamic simulations that can take weeks or months to complete for a single set of parameters. Most previous chemical models simply neglected the wind altogether, treating irradiated disks as if they were just isolated disks bathed in a stronger UV field.

This simplification raises an important question: does the wind actually matter for disk chemistry? After all, the wind is comparatively tenuous compared to the dense disk interior. Perhaps we could safely ignore it.

Some Interesting Results

To answer this question, we post-processed existing hydrodynamical simulations with a detailed thermochemical code, comparing models that included the full wind structure against those where we artificially removed all the wind material. The results were striking.

First, we discovered an unexpected heating mechanism. While UV radiation is rapidly attenuated at the disk surface in both wind and windless models, the wind itself acts as a reprocessing medium. It absorbs UV photons and re-emits them as infrared radiation, which has much lower opacity and can penetrate deep into the disk interior. This raises midplane temperatures by several Kelvin... modest, but sufficient to shift the balance between freeze-out and thermal desorption for volatile species like CO. In the outer disk regions where CO would normally be frozen onto grain surfaces, this extra heating can keep it in the gas phase, increasing abundances by orders of magnitude in localised regions.

Second, when we generated synthetic observations, predicting what telescopes like ALMA and JWST would actually see, the differences were even more dramatic. Molecular line fluxes varied by factors ranging from a few up to 100,000 between wind-inclusive and wind-free models. Crucially, this wasn't primarily because abundances changed within the disk itself. Instead, the wind directly contributes substantial emission, sometimes dominating the total flux.

The PUFFIN Framework

These findings highlighted a problem: we needed to study chemistry across wide parameter ranges (different stellar masses, disk sizes, UV field strengths) to understand when and how these effects matter. But we couldn't run hundreds of radiation-hydrodynamic simulations—that would take decades of computing time.

Our solution was to develop PUFFIN (Python Utility For FUV Irradiated disk deNsities), a parametric framework that generates physically-motivated disk-wind density structures in seconds to minutes rather than weeks to months. We calibrated it against over 600 hydrodynamical simulations, enabling it to reproduce the essential morphological features, including the photoevaporative wind.

With this tool in hand, we could finally explore the chemistry systematically. Applying it to CO across a comprehensive parameter grid revealed some interesting trends. External UV irradiation acts as a second-order effect on disk chemistry, with its importance depending critically on stellar mass. For massive stars (≥3 solar masses), stellar heating dominates completely and external fields barely matter. But for lower-mass stars (≤0.6 solar masses), external fields of 1000 G₀ or stronger can fundamentally alter the CO gas-ice balance across much of the disk.

Broader Implications

These results have important implications for interpreting observations. Recent ALMA surveys of Orion Nebula disks have found puzzlingly high CO abundances and gas-to-dust ratios. Our models suggest that wind-driven infrared heating could contribute to these observations by keeping volatiles in the gas phase. Meanwhile, JWST spectroscopy of irradiated disks is revealing diverse chemical signatures that don't always match predictions from simpler models—and our work suggests that properly accounting for wind structure and geometry may be essential for understanding these observations.

More fundamentally, this work challenges the traditional approach of treating externally irradiated disks as isolated disks plus a UV field. The wind isn't just a mass-loss mechanism, it's an integral component of the disk's physical and chemical structure, affecting temperatures, abundances, and observables in interconnected ways.

Future Directions

The PUFFIN framework enables systematic searches for molecular tracers that could diagnose external irradiation in observations where we can't spatially resolve disk and wind components. We're also extending beyond CO to other crucial volatiles (water, organics, sulfur-bearing species, nitriles etc), to understand how external irradiation shapes the complete compositional inventory available for planet formation. Finally, we're working toward connecting disk chemistry to planetary atmospheres: can the radial variations we see in volatile abundances leave detectable fingerprints in exoplanet compositions that tell us about their formation environment?

The broader goal is to build toward a truly comprehensive understanding of how environment shapes planet formation, moving beyond the study of isolated disks to encompass the diverse conditions under which most planets actually form.

This research was originally published in Monthly Notices of the Royal Astronomical Society:

• Keyte & Haworth (2025) "Impact of photoevaporative winds in chemical models of externally irradiated protoplanetary discs"
• Keyte & Haworth (2026) "A parametric model for externally irradiated protoplanetary disks with photoevaporative winds"