What artificial photosynthesis means

Artificial photosynthesis uses sunlight to drive chemistry that stores energy in molecular bonds — splitting water into hydrogen and oxygen, or reducing carbon dioxide into carbon-based fuels. The appeal over photovoltaics plus batteries is storage and transport: a fuel keeps, moves, and drops into existing infrastructure. The appeal over biology is efficiency, since natural photosynthesis converts only a small fraction of incident sunlight into stored chemical energy.

Where the science stands

Established Water splitting is the solved half. Laboratory tandem photoelectrochemical devices have reached solar-to-hydrogen efficiencies around 14%, and in 2025 a module-sized perovskite-based “artificial leaf” — a 16 cm² array of sixteen one-centimetre sub-cells — reported above 10% while explicitly targeting durability and scalability alongside efficiency. Ten per cent has long been the field's stated practical threshold.

Established Efficiency falls with area, sharply. A standalone bismuth-vanadate–silicon artificial leaf achieving 8.4% at small scale delivered 2.7% when scaled to 441 cm² under natural sunlight. This is the field's defining engineering problem and it recurs across chemistries: resistive losses, illumination non-uniformity and bubble management all worsen with size.

Frontier Reducing CO2 is much harder than splitting water, because of product selectivity, electrolyte constraints and higher voltage requirements. Systems pairing photovoltaics with catalysts have reached solar-to-carbon-monoxide efficiencies above 6.5%; fully monolithic photocatalyst sheets converting CO2 to formate have achieved excellent selectivity at solar-to-fuel efficiencies below one tenth of one per cent. The gap between those two numbers is the gap between an integrated device and a wired system.

Frontier The largest outdoor demonstration reported covered about 103 square metres, producing carbon monoxide via photothermal reverse water-gas shift at over 16% solar-to-chemical efficiency across a day, with daily output on the order of tens of cubic metres. That was roughly sixteen times larger than the previous large-scale demonstration — genuine progress, and an area smaller than a tennis court.

Speculative Durability and cost remain unresolved together: the highest efficiencies often come from materials or architectures that are expensive, short-lived, or both. Handwave Nothing in this field is near commercial deployment, and framing artificial photosynthesis as a near-term carbon solution misrepresents a laboratory science.

The binding bottleneck

Established The constraint is the scale-efficiency trade — every reported record must be read together with its active area, and the field's honest metrics are efficiency, area and operating hours reported jointly. Frontier Second is durability, since photoelectrodes degrade in the electrolytes they must operate in. Frontier Third, for the CO2 route, is selectivity at useful current densities. Frontier Fourth is the comparison nobody in the field can avoid: photovoltaics plus electrolysis is an unglamorous, commercially available alternative that integrated devices must beat on cost, not on elegance.

Ethics and governance

Established The direct ethical stakes are low; this is early-stage research with no deployment footprint. Frontier The governance issue is claim discipline: over 95% of global hydrogen production still comes from fossil fuels, and “solar fuel” language is used loosely enough that buyers cannot always tell what they are purchasing. Frontier Materials sourcing — iridium, platinum group metals, lead in some perovskites — would become a real constraint and a real environmental question at any meaningful scale.

Timelines

  • 10 yr: Frontier square-metre modules with durable double-digit solar-to-hydrogen efficiency; CO2 routes still well behind.
  • 25 yr: Speculative pilot-scale solar fuel production, competing against photovoltaics-plus-electrolysis rather than against fossil fuels.
  • 50 yr: Speculative material contribution to fuel supply, conditional on cost and durability advances not currently in view.
  • 100 / 250+ yr: Speculative closing the carbon cycle industrially, which is the version of this idea that would matter.