Artificial photosynthesis

In plain language

An artificial leaf is a device that uses sunlight to split water into hydrogen and oxygen. Light-absorbing materials create electrical charges, and catalysts on the surfaces use those charges to drive the reactions that separate water. Some designs connect the light absorber to a separate solar cell. More integrated designs combine both jobs in a compact unit with no external wiring.

The word photosynthesis is a comparison, not a claim of identical chemistry. Plants store energy in carbon compounds, whereas these devices aim to make hydrogen and oxygen. A good result in the laboratory does not show that a device lasts for years, that the hydrogen is purified and stored safely, or that the whole system has a low carbon footprint. Most results so far come from small prototypes tested under controlled light.

Technical explanation

Solar-to-hydrogen (STH) efficiency is the chemical energy of the hydrogen produced, divided by the incident solar energy, under a stated test method. Reported values depend on device area, illumination spectrum, temperature, electrolyte, how the gas is measured, and whether the device needs an external bias voltage. Unbiased operation means the device needs no applied voltage to drive the reaction. An early wireless silicon-based design, reported by Reece and colleagues in 2011, used a triple-junction amorphous silicon photovoltaic, an earth-abundant alloy catalyst for hydrogen evolution and a cobalt-borate catalyst for oxygen evolution. It reached 4.7% STH when wired and 2.5% in the wireless configuration under one-sun AM1.5 simulated sunlight, near neutral pH.

Two 2025 studies in Nature Communications show the current state of the field. Hansora and colleagues reported a 16 cm2 module built from a 4 by 4 array of 1 cm2 photoelectrodes. The light absorber was chlorine-doped formamidinium lead triiodide, a lead-halide perovskite, and the module reached an unbiased STH efficiency of 11.2% under one-sun illumination. The abstract reports high stability for 140 hours for the 1 cm2 photoelectrodes. Separately, Liu and colleagues reported a bismuth vanadate photoanode paired with a silicon photovoltaic cell. The standalone device reached 8.4% STH in a laboratory test, and the photoanode was stable for over 520 hours under AM 1.5G illumination. A scaled-up leaf of 441 cm2 reached 2.7% STH in natural sunlight. Each study uses different materials, areas and test conditions, so the figures are not a controlled comparison.

This profile covers water splitting only. Reducing carbon dioxide to carbon-based fuels is a separate reaction with its own carbon source, product mixture and separation problems, and the sources reviewed here do not show it at a comparable stage.

What the evidence says

Each statement is typed by the strength of its evidence and linked to the source that supports it.

Limitations and open problems

• The 11.2% module result is for a 16 cm2 module under one-sun illumination. The 140-hour durability result is for 1 cm2 photoelectrodes, not the full module, and no outdoor field trial of the module is reported in the sources reviewed.
• The 8.4% and 2.7% figures come from different devices, areas and illumination conditions, so they are not directly comparable. The 2.7% figure is for a scaled-up leaf tested in natural sunlight. The sources do not show that a small device's efficiency carries over to a larger one.
• The best-performing absorber in these sources contains lead. Encapsulation in the reported study does not establish whole-device safety, lead containment after damage, or recycling at end of life.
• Hydrogen is not automatically low-carbon. The lifecycle impact depends on how the absorber, catalysts and enclosure are made and how long they last, and on how the hydrogen is compressed, stored and moved. The only lifecycle claim in the sources is the authors' own and has not been independently checked.
• The sources cover hydrogen from water only. They do not show that artificial leaves can make carbon-based liquid fuels from carbon dioxide.

Possible future uses

Product concepts: Solar hydrogen roof tile

Timeline

1. 4 Nov 2011Reece and colleagues report a wireless silicon-based device reaching 2.5% solar-to-hydrogen efficiency under one-sun AM1.5 simulated sunlight, and 4.7% in a wired configuration. Sources: [3]
2. 21 Mar 2025Liu and colleagues report a standalone bismuth vanadate-silicon artificial leaf with 8.4% solar-to-hydrogen efficiency, and 2.7% for a scaled-up leaf of 441 cm2 in natural sunlight. Sources: [2]
3. 6 May 2025Hansora and colleagues report a 16 cm2 perovskite-based artificial leaf module with 11.2% unbiased solar-to-hydrogen efficiency under one-sun illumination. Sources: [1]

Futures this could help bring about

How far away?

Making fuel from sunlight like a leaf

Engineering prototypes · 5 of 8 steps done

Glossary

Artificial leaf

An engineered device that uses light to drive chemical reactions, often water splitting. The name describes the function, not plant biology.

Photoelectrochemical (PEC) device

A system in which light-absorbing electrodes and electrochemical reactions are combined in an electrolyte.

Solar-to-hydrogen (STH) efficiency

The fraction of incident solar energy that is stored in the hydrogen produced, as defined by the reported test method.

Unbiased operation

Operation without an applied external voltage to drive the reaction. It does not mean the wider plant needs no other equipment.

Sources

1. peer reviewed paper
   Scalable and durable module-sized artificial leaf with a solar-to-hydrogen efficiency over 10%
   Dharmesh Hansora, Rashmi Mehrotra, Eunseo Noh et al.. Nature Communications, 6 May 2025. DOI: 10.1038/s41467-025-59597-2. Accessed 21 Oct 2026.
2. peer reviewed paper
   A standalone bismuth vanadate-silicon artificial leaf achieving 8.4% efficiency for hydrogen production
   Boyan Liu, Xin Wang, Yingjuan Zhang et al.. Nature Communications, 21 Mar 2025. DOI: 10.1038/s41467-025-58102-z. Accessed 21 Oct 2026.
3. peer reviewed paper
   Wireless solar water splitting using silicon-based semiconductors and earth-abundant catalysts
   S. Y. Reece, J. A. Hamel, K. Sung et al.. Science, 4 Nov 2011. DOI: 10.1126/science.1209816. Accessed 21 Oct 2026.
4. review article
   Solar Water Splitting Cells
   M. G. Walter, E. L. Warren, J. R. McKone et al.. Chemical Reviews, 10 Nov 2010. DOI: 10.1021/cr1002326. Accessed 21 Oct 2026.

Revision history

3 changes since this profile was created.