Photosynthesis is the process by which plants, algae and certain bacteria convert carbon dioxide and water into sugars using energy drawn from sunlight. It sustains almost all life on earth, releases the oxygen we breathe, and takes place inside structures so small that a single leaf contains them in their billions. Its chemistry was largely unknown until the twentieth century, and important details were worked out only in the past three decades. Understanding these details has consequences for climate science, for agriculture and, increasingly, for attempts to build devices that imitate what plants do.

The basic equation of photosynthesis is simple. Six molecules of carbon dioxide and six of water, in the presence of light and chlorophyll, yield one molecule of glucose and six of oxygen. The elegance of the equation hides a far more intricate process carried out in two linked sets of reactions. In the first, chlorophyll captures the energy of a photon and uses it to split water into protons, electrons and oxygen. This is the step that releases the oxygen plants eventually give to the atmosphere. In the second, the electrons are passed through a series of carriers, and their energy is used to fix carbon dioxide into sugars by an enzyme called RuBisCO.

The water-splitting step is particularly striking. A single cluster of four manganese atoms, along with a calcium atom, accomplishes a reaction that human chemists have only recently come close to imitating. The cluster sits inside a large protein complex known as Photosystem II, embedded in the membranes of chloroplasts. Biochemist Dr. Aysha Khatun, working at Imperial College London, has described the cluster as 'the smallest factory in the world', because it operates continuously in daylight, withstanding the chemically aggressive environment of its own products. Its structure was finally solved to atomic precision by X-ray crystallography in 2011.

RuBisCO, the enzyme that fixes carbon dioxide, is almost equally remarkable but in a different way. It is the most abundant protein on earth, present in large quantities in every green leaf. Unfortunately, it is also slow and imprecise. RuBisCO was evolved in an atmosphere with far less oxygen than today, and about a quarter of the time it mistakenly binds oxygen rather than carbon dioxide, a wasteful reaction that plants have to correct with additional energy. Many attempts to improve crop productivity have focused on modifying RuBisCO, either by introducing variants from algae that work faster or by helping the enzyme encounter carbon dioxide at higher local concentrations.

Some plants have evolved their own responses to RuBisCO's imperfection. In hot, dry climates, maize and sugar cane use a mechanism called C4 photosynthesis, in which carbon dioxide is concentrated in specialised cells before it reaches RuBisCO. C4 plants are substantially more efficient than standard C3 plants in high temperatures, and a major research effort funded by the Bill and Melinda Gates Foundation has attempted to transfer C4 photosynthesis into rice, a vital staple that normally uses the less efficient pathway. Progress has been slow but steady; fully photosynthetic C4 rice remains a long-term goal rather than an achievement.

Artificial photosynthesis is another active area. Several laboratories have built devices that use sunlight to split water into hydrogen and oxygen, using catalysts inspired by the manganese cluster of Photosystem II. The hydrogen produced is in principle a clean fuel; the difficulty is that the catalysts degrade faster than their biological equivalents, and the overall efficiency of artificial systems remains lower than that of plants. Dr. Khatun's team has studied how natural Photosystem II repairs itself continuously, replacing damaged components, and has suggested that artificial systems will need similar self-repair mechanisms before they can be deployed at scale.

The public relevance of this work is wider than it first appears. Agriculture has relied for millennia on the efficiency of natural photosynthesis; now, as populations rise and climates shift, even small improvements in photosynthetic efficiency could have large effects on food supply. Meanwhile, artificial systems that produce hydrogen or fuels from sunlight and water offer a possible route to low-carbon energy storage. Both directions draw on the same fundamental science. Dr. Khatun has observed that the molecules at the heart of photosynthesis seem, on first acquaintance, implausibly well-designed; the closer research looks, however, the more the design resembles a long list of workable compromises, and the more useful it becomes as a guide to what an engineered system would need to do.