How sunlight-powered molecules can trap and release carbon dioxide

Introduction

The idea of using sunlight to directly capture carbon dioxide sounds like science fiction — but a team led by Richard Liu at Harvard has turned it into a reproducible laboratory reality. In a paper published online August 13, 2025, researchers described a new class of light-activated molecules, called fluorenol "photobases," that use visible sunlight to make aqueous solutions temporarily more basic, bind CO2 from air, and then release it again in the dark for collection (Purdy et al., Nature Chemistry 2025: https://doi.org/10.1038/s41557-025-01901-0).

This article explains how the chemistry works, highlights related independent studies, and explores why this approach could change how we think about direct air capture (DAC).

what is a photobase (and how is it different from a photoacid)?

A photobase is a molecule whose basicity — its tendency to accept protons — increases when it absorbs light. In plain language: the molecule becomes more likely to pull hydrogen (and thus increase pH) while it is excited by photons. That pH shift can generate hydroxide in water, which then reacts with dissolved CO2 to form carbonate or bicarbonate and trap CO2. This is the chemical mirror to photoacids, which become more acidic under illumination and have been used to drive other light-controlled reactions (definitions and mechanistic background: Purdy et al., ChemRxiv preprint: https://chemrxiv.org/engage/chemrxiv/article-details/67dd83ef81d2151a0253f19b).

how the Harvard fluorenol photobases work

• Under visible-light irradiation the fluorenol derivatives undergo an excited-state process that increases their pKa by almost 6 units, producing hydroxide in water.
• The hydroxide rapidly converts dissolved CO2 into carbonate/bicarbonate, concentrating CO2 in solution.
• When the light is turned off and the molecules relax to their ground state, basicity falls back, and CO2 is released in a form that can be collected.

Key reported metrics include a photochemical quantum yield above 20% for the hydroxide-release step and low cycle degradation (~1% per cycle under test conditions), plus demonstrable multi-cycle capture from ambient air using sunlight (Purdy et al., Nature Chemistry 2025: https://doi.org/10.1038/s41557-025-01901-0).

independent corroboration and alternative molecular routes

This Harvard work is part of a broader, convergent trend. Other groups have reported related sunlight-driven CO2-capture schemes that use different molecular tricks:

• Photo-DAC preprint: A guanidine-based photobase (PyDIG) that isomerizes under UV light to a more basic form and captures CO2 without a heating step. Reported cyclic capacities were roughly 0.21–0.26 mol CO2 per mol solvent and stable cycling for several runs (Photo‑DAC, ChemRxiv Apr 2025: https://chemrxiv.org/engage/chemrxiv/article-details/67e6a2f3fa469535b9cc2d69).
• Cornell/Chem paper: A sunlight-powered scheme inspired by plant chemistry that uses a photoenol to bind CO2 and a light-driven release step; importantly, it demonstrated operation with real flue-gas samples and contaminants (Cornell news summary, May 2025: https://news.cornell.edu/stories/2025/05/first-system-uses-sunlight-power-carbon-capture).

Together, these studies show multiple molecular strategies (fluorenol photobases, guanidine photobases, photoenol capture chemistry) and different spectral requirements (visible vs UV). They point to the same big idea: use photons instead of heat for the capture/release energy input.

why this is important: energy, cost, and scale

Traditional sorbent-based DAC systems typically use heat to desorb CO2, which is energetically costly and often requires fossil-derived or high-grade heat. Light-driven molecular cycles promise several potential advantages:

• direct use of solar energy as the regeneration input, reducing thermal energy demand (Purdy et al., Nature Chemistry 2025: https://doi.org/10.1038/s41557-025-01901-0);
• operation at ambient temperature, which simplifies equipment and may reduce capital costs (Photo‑DAC preprint; Cornell work);
• modularity and possible deployment in distributed systems if molecules and processes can be engineered for robustness and low cost.

However, important scale-up challenges remain: solubility limits, long-term photostability in real atmospheres (oxygen, particulates), wavelength tuning to match the solar spectrum, and integrating capture, concentration, and storage steps into a single practical process (Purdy et al., ChemRxiv preprint: https://chemrxiv.org/engage/chemrxiv/article-details/67dd83ef81d2151a0253f19b).

case study: what the lab demonstrations show

The Harvard team demonstrated ambient-air CO2 uptake cycles under natural sunlight and sunlight simulators, measured a high quantum yield for hydroxide generation (>20%), and reported multi-cycle operation with low degradation (~1% per cycle) in bench-scale experiments (Purdy et al., Nature Chemistry 2025: https://doi.org/10.1038/s41557-025-01901-0). Independently, the Photo‑DAC and Cornell groups showed similar ambient-temperature cycling and operation on real emission streams, reinforcing that multiple molecular approaches can work outside tightly controlled conditions (Photo‑DAC preprint: https://chemrxiv.org/engage/chemrxiv/article-details/67e6a2f3fa469535b9cc2d69; Cornell news: https://news.cornell.edu/stories/2025/05/first-system-uses-sunlight-power-carbon-capture).

why this matters

• Could solar-driven capture lower the energy penalty of removing CO2 from air?
• Might distributed, low-temperature DAC become economically practical if sunlight replaces heat for sorbent regeneration?

These questions are not rhetorical. If photochemical cycles can be made robust, inexpensive, and compatible with large-scale chemical handling, they could open new pathways for removals that complement afforestation, industrial carbon capture, and emission reductions.

questions to ponder

• How will molecules be engineered to survive years of sunlight, oxygen, and dust in real-world deployment?
• What is the best way to couple light-driven capture to long-term CO2 storage or utilization systems?
• Can process engineers design reactors that efficiently capture sunlight and move CO2 from dilute air to concentrated streams at scale?

next steps and what to watch for

Researchers and companies will need to focus on:

• improving photostability and solubility of photoactive molecules;
• shifting absorption into the strongest parts of the solar spectrum (visible and near-IR);
• process demonstrations at pilot scale that include real air or flue gas streams and full capture-to-release mass balances.

Conclusion: a practical revolution in the making

The Harvard fluorenol photobase work, together with independent Photo‑DAC and Cornell studies, shows a real, reproducible path toward sunlight-powered CO2 capture (Purdy et al., Nature Chemistry 2025: https://doi.org/10.1038/s41557-025-01901-0; Photo‑DAC preprint: https://chemrxiv.org/engage/chemrxiv/article-details/67e6a2f3fa469535b9cc2d69; Cornell news: https://news.cornell.edu/stories/2025/05/first-system-uses-sunlight-power-carbon-capture). These studies do not promise an immediate drop-in replacement for existing DAC plants, but they do offer a revolutionary concept: use photons — abundant, renewable, and free — to do the heavy lifting of sorbent regeneration.

Concrete possibilities include off-grid modular DAC units powered by sunlight, lower-operating-cost capture at industrial sites with abundant solar exposure, and hybrid systems that combine thermal and photochemical regeneration to balance performance across climates.

If you care about climate solutions, a practical question follows: will researchers, funders, and industry move quickly enough to translate promising lab chemistry into pilot plants? The next few years of scale-up experiments will tell us whether sunlight-driven molecules become a niche lab curiosity or a new workhorse technology for atmospheric CO2 removal.

Original source: https://doi.org/10.1038/s41557-025-01901-0