ASK a biologist and a renewable energy researcher to come up with a list of the most important chemical reactions in the world, and the chances are their lists would not have much in common. But one reaction would undoubtedly be near the top of both: the light-driven splitting of water molecules into oxygen, hydrogen ions and electrons.
This is the heartbeat of photosynthesis, the process by which plants, algae and cyanobacteria capture energy from sunlight and fix it into sugars. Both researchers would have good reason to choose it. The biologist would point out that the reaction supports all complex life on Earth, supplying it with both energy and oxygen. The energy researcher would say that it might just hold the key to the world’s energy problems.
Photosynthesis is the most successful solar converting mechanism on Earth. And when nature has invented such a successful system, it would be foolish to ignore it as a potential source of renewable energy, says Stenbjörn Styring, professor of biochemistry at Lund University in Sweden. Artificial photosynthesis would make it possible to harness sunlight to produce limitless quantities of hydrogen or other energy-rich fuels from water, cleanly and cheaply.
Advertisement
The problem is, how plants perform that crucial reaction of splitting water molecules is a mystery. According to electrochemical theory, the energy required to dismember water is more than enough to destroy any biological molecule. Yet plants do it all day, every day, without any ill-effects. Until we know how they achieve this, there seems little chance of repeating the feat in a test tube.
But now a team at Imperial College London led by Jim Barber and So Iwata has made what could be the decisive breakthrough. They have worked out the precise spatial arrangement of a small number of metal ions, oxygen atoms and water molecules at a crucial site in the plant’s photosynthetic machinery where water splitting occurs – the so-called “catalytic core”. Many leading researchers in the field now agree that the debate over various competing structures has been all but settled. And with the structure in the bag, the chemical mechanism of water splitting should quickly follow. “It’s a great bit of work, and, yes, represents a major breakthrough,” says Fraser Armstrong, an inorganic chemist at the University of Oxford.
This may be just what renewable energy researchers have been waiting for. Knowing the structure of the catalytic core will provide a springboard for research into artificial photosynthesis, says Bill Rutherford, a structural biologist at France’s Atomic Energy Commission in Saclay, France. “There has been a very long wait for the structure of the site, so there is euphoria and hype breaking out all over,” he says. “Hype or no hype, I am certain that the structure is the beginning of the next big phase in the subject.” This next phase is to unravel the precise chemical events involved in splitting water molecules. “We want to understand this unique chemistry because we can’t reproduce it as yet artificially,” says Barber.
Natural photosynthesis takes place in two large assemblies of proteins, metal ions and green pigments called chlorophylls that sit cheek-by-jowl inside the chloroplast, the plant’s photosynthetic apparatus. Chemists who are trying to build artificial water-splitting systems are interested in the assembly called photosystem II (so called because it was discovered second).
When a photon of light strikes photosystem II, it is channelled into a specialised chlorophyll molecule called P680. This releases a high-energy electron which is sent on a circuitous route before being used to reduce carbon dioxide to sugars. P680 then returns to its ground state to await another photon strike. But before it can do so it needs to replace its lost electron. This is where the catalytic core comes in: for every electron lost by P680, the catalytic core pulls one out of a water molecule to replenish it. After four rounds of electron transfer the catalytic core spits out a molecule of oxygen and four hydrogen ions, and reloads. Overall, light converts two molecules of water into a molecule of oxygen, four electrons and four hydrogen ions.
Constructing an artificial system that can operate continuously in a similar way poses three basic problems, Styring says. Two have already been solved. Researchers know how to capture the sun’s energy, and they can also transmit it, in the form of electrons, to a reaction centre to produce hydrogen. But what they have been unable to do until now is complete the cycle and replenish the electrons stripped out by sunlight. Without this vital third stage, the whole process quickly grinds to a halt. “This is the difficult part and is a long-term project,” Styring says.
In plants these replenishing electrons come from the water-splitting reaction. If artificial photosynthesis researchers could mimic this crucial step, it would open the way to converting sunlight into usable energy in the same way plants do. “This would provide a never-ending, environmentally friendly starting substance,” Styring says. And he believes that Barber’s and Iwata’s discovery of the catalytic core’s structure could at last make this possible, though he estimates it could take 10 years to replicate the natural water-splitting chemistry in an artificial system.
Photosynthesis researchers have long been aware that the secret of water splitting is somehow embodied in the structure of the catalytic core. Although there are variations between species in the protein components of photosystem II, the catalytic core appears to be the same in all plants, algae and cyanobacteria. This suggests that the arrangement of atoms is critical to its function. Change it in any way and it loses its powers. So understanding the structure of the catalytic core has been seen as an important goal, with at least three research teams chasing it.
The first breakthrough came in 2001, when a team led by Petra Fromme of the Technical University of Berlin in Germany published the first high-resolution structure of photosystem II obtained by X-ray crystallography (Nature, vol 409, p 739). This was no easy task, because some of the key proteins in the complex are in a constant state of repair from oxidative damage. Then in 2003, a team at the Harima Institute, a structural biology research centre near Kobe, Japan, published an even clearer X-ray structure at a slightly higher resolution (Proceedings of the National Academy of Sciences, vol 100, p 98).
These structures proved beyond doubt that the catalytic core comprises four manganese ions, a calcium ion, several oxygen atoms and at least two water molecules, all held in place by a protein scaffold. But they still did not reveal the precise spatial geometry of the metal atoms. The prevailing view was that the centre of the catalytic core was made up of four manganese ions and four oxygen ions arranged in a cube, with a single calcium ion just outside. But the precise arrangement of the manganese ions, and whether or not the calcium was inside the catalytic core or just outside it, remained unknown.
Barber and his team have now resolved those issues in an ingenious analysis of X-ray diffraction images of the catalytic core of the cyanobacterium Thermosynechococcus elongatus. What they found came as a real surprise. Originally, they held to the mainstream view that the catalytic core consisted of four central manganese ions with a calcium ion tagging along. But their final structure, obtained in December 2003 and published in March (Science, vol 303, p 1831), suggests something different: the catalytic core is actually a distorted cube comprising three manganese ions plus a calcium ion, interconnected by oxygen atoms. The fourth manganese ion sits outside, with a water molecule bound to it (see Diagram). Most researchers agree with their analysis. “The new data look good for this structure,” says Gary Brudvig of Yale University, who is a pioneer of the chemistry of water splitting.
The findings don’t just pin down the structure. They also have important implications for understanding – and copying – the chemistry of water splitting. “Certainly, several mechanistic proposals in the literature can be thrown in the bin,” says Styring.
Attempts to elucidate the chemistry go back more than three decades. The first major breakthrough came in 1969, when Bessel Kok of Martin Marietta Laboratories in Baltimore, Maryland, and Pierre Joliot of the Institute of Physico-Chemical Biology in Paris developed a model in which the catalytic core goes through a four-step cycle. Starting from the observation that it takes four photons to generate one molecule of oxygen, Kok and Joliot proposed that the catalytic core starts at a resting state called S0, then moves through four successive states – S1, S2, S3 and S4 – in response to the absorption of four photons of light by the photosystem. As this sequence proceeds, the catalytic core accumulates enough “electron-stripping power” or redox potential to extract electrons from water (see “Redox potential”). After the fourth step, two water molecules are split into a molecule of oxygen plus four electrons and four hydrogen ions, and the catalytic core returns to S0. This model has prevailed ever since and is now known as the S-state cycle or “Kok clock”.
The detailed chemistry of each step in the cycle remains unknown, though numerous mechanisms have been suggested. Kok himself believed that the catalytic core dismembered the water molecules bit by bit, extracting one electron at a time, while others have suggested slightly different schemes. But according to Barber the new structure means that they can all be ruled out. “I don’t believe that’s thermodynamically possible,” he says.
Barber is now convinced that the critical chemistry required to initiate the splitting of water happens only after the third step of the S-state cycle, and that it takes place at the single manganese ion outside the distorted cube. His argument starts from the fact that the redox potential of water is +2.5 volts, yet each step of the S-state cycle raises the catalytic core’s redox potential by only +1 volt. He therefore proposes that, as the catalytic core loses electrons, redox potential accumulates on the isolated manganese ion. At S0 this ion has a redox potential of about 0; by S3 it has built up to about +3 volts, sufficient to mount an attack on water.
At this point things start to get interesting. First, the isolated manganese ion steals an electron from the water molecule bound to it, consuming about 2.5 volts of redox potential and leaving a hydroxyl (OH) radical and a hydrogen ion. The catalytic core then clicks through the fourth step of the cycle, raising the manganese ion’s redox potential sufficiently to mount a further raid, this time on the OH radical, consuming about another 1.5 volts of redox potential. This creates a highly reactive oxygen atom and a second hydrogen ion.
At this point the calcium ion within the distorted cube joins the fray. It, too, has a water molecule bound to it, held in just the right place to be instantly attacked by the oxygen atom. This final reaction produces a molecule of oxygen, two further protons and two electrons. This last is a “downhill” reaction returning 0.6 volts to the system.
Other researchers agree that Barber’s proposed mechanism is consistent with his structure, but some point out that other schemes are also possible. These cannot be finally ruled out until more detailed X-ray diffraction images are available to reveal more accurate structures for the core.
Styring and other leading researchers expect this to happen very soon. And they are convinced that it will then be possible to mimic the chemistry of photosystem II in an artificial system. The key challenge is to replicate the S-state cycle in which water molecules provide a continuous supply of electrons. Attempts so far, by Styring and physical chemist Leif Hammarström of Uppsala University in Sweden, for example, used manganese in their “catalytic core”, and replaced the relatively fragile chlorophyll-protein complexes of photosystem II with ruthenium, for light capture, and iron, which acts as the reaction centre.
In the latest devices, a cluster of ruthenium atoms captures a photon of light and donates a high-energy electron to an iron reaction centre, which uses the electron to extract a hydrogen ion from water to make a hydrogen atom. A complex of manganese ions then replenishes the ruthenium’s electron, and the reaction happens again. But at this point the system grinds to a halt, as there is nothing to replace the electron from the manganese. The challenge now is to close the loop, drafting in the water-splitting chemistry of photosystem II to replenish the manganese cluster’s lost electrons. “It is here we need all we can find out about photosystem II,” says Styring.
There is likely to be one crucial difference between artificial photosynthesis and its natural counterpart: it will produce hydrogen. Once two electrons have been extracted, instead of being retained as a source of energy for production of sugars, they would react with two hydrogen ions, also obtained from water splitting, to form a molecule of hydrogen. This, says Styring, is the easy part, and should be working in a year or two.
The emergence of photosynthetic water splitting was a pivotal event in the evolution of life on Earth, creating the conditions for multicellular life to exist. Now, 2.5 billion years on, human ingenuity is struggling to repeat the feat in an effort to provide a truly sustainable source of energy.
Redox potential
Redox potential is a measure of a chemical compound’s affinity for electrons; it is measured in volts. Chemicals that give their electrons away easily have a negative redox potential. Electron-loving chemicals, have a positive redox potential. The higher the number, the more strongly they attract electrons – and the more energy is required to rip an electron from their grasp. Water’s redox potential is high, about +2.5 volts. Only chemicals entities with an even higher redox potential, such as oxygen atoms, can extract electrons from water.
