Sunlight through jungle like vegetation

Artificial photosynthesis: Mimicking nature's most powerful trick

There’s wind, there are photovoltaic cells, but we need a way to store the energy. Could artificial photosynthesis - turning the sun’s energy into a fuel, such as hydrogen - one day replace dirty coal and other fossil fuels as a plentiful source of green, clean energy? We analyse how far this technology has come.

“If the leaf can do it, we can do it even better.”

It was a valiant statement, made six years ago at the opening of the ‘Solar to Fuels and Back Again’ symposium at Imperial College London. Numerous conferences and symposiums and plenty of research later, however, we are still not able to mimic the most efficient power station created by nature, the humble leaf.

The proponents of artificial photosynthesis make a strong case for our need of a breakthrough. Fossil fuels keep pumping unprecedented levels of carbon dioxide into the atmosphere, and we are not anywhere close to meeting pollution targets to contain climate change. The use of renewable energy is growing, with hydro dams and wind turbines popping up here and there; we even use the sun, with more and more photovoltaic solar parks dotting the landscape.

However, the wind doesn’t always blow; hydropower has environmental side-effects; and photovoltaics stop working as soon as the sun drops below the horizon.

Artificial photosynthesis, in contrast, hopes to mimic what nature has been doing all along: convert sunlight directly into chemical fuels, which would make it possible to simultaneously harvest and store energy from the most abundant and sustainable energy source available. “If we can make this work efficiently and cheaply on a global scale, we will have truly solved the energy problem,” says Roel van de Krol, a materials scientist at the Helmholtz-Zentrum Berlin and the Technische Universität Berlin.

If we then use the direct product of this conversion, hydrogen, to convert CO2 into another fuel (for example methanol), we’d be on our way into carbon-neutral future.

“To come up with a solution of using sunlight to create energy which then runs our everyday life is critical for this century. If we don’t, we may be doomed as a society,” says Uwe Bergmann, a physicist at SLAC National Accelerator Laboratory in Menlo Park, California. “It’s mind-boggling that you have a leaf, a small leaf, able to use sunlight at room temperature and ambient pressure to produce oxygen and fix carbon. When you try to do it yourself, you can only wonder how amazing it all is.”

In nature, it works like this: a leaf harvests energy from sunlight and uses it to split water into oxygen, protons and electrons. Plants then combine the protons and electrons with CO2, converting it into carbohydrates, or sugar, to store the chemical energy that living cells use.

To achieve the same process artificially, scientists want to use sunlight to efficiently split water and produce a fuel that can be stored. First, a photovoltaic material captures photons and generates electrons; next a chemical catalyst breaks the chemical bonds of water’s hydrogen and oxygen atoms.

The resulting protons and electrons are then combined into molecular hydrogen, which can be used as a fuel. Photochemical CO2 reduction is a separate reaction that yields fuel as well, in the form of hydrocarbons such as methane. Most of the energy stored comes from the water splitting, and the rest from the carbon fixation. So in the end, efficient artificial photosynthesis would only rely on water, sunlight and CO2 to produce energy, without any harmful by-products. But to be commercially viable, such a system also has to be efficient, stable, safe and inexpensive.

Scientists started toying with the idea of artificial photosynthesis and solar fuels after a Japanese chemist, Akira Fujishima at the University of Tokyo in 1967, came across the photocatalytic properties of titanium dioxide and realised that they could be used for hydrolysis. It took another 25 years before the research started to receive more serious attention. In 1994, scientists from the universities of Lund, Uppsala and Stockholm in Sweden established the Swedish Consortium for Artificial Photosynthesis, the first of its kind.

Since then, more and more teams have started to pop up around the world, all hoping to crack the secret of producing cost-effective solar fuel. In 2010, the US government invested $122m to set up the country’s largest research programme devoted to developing artificial solar fuel generation technology - the Joint Center for Artificial Photosynthesis (JCAP). Its two main centres are at the California Institute of Technology and the Lawrence Berkeley National Laboratory, alongside a number of partner institutions.

To make an efficient, cheap, durable, and scalable solar fuel device, scientists first have to develop at least three key components: a light absorber, a catalyst to trigger the reduction of water to hydrogen, and a catalyst for the oxidation reaction. The light absorber would convert light into energetic electrons and holes, while the catalysts have to make sure that these charge carriers can be effectively used to drive the electrochemical reactions that convert cheap resources such as water and/or CO2 into chemical fuels such as hydrogen and hydrocarbons.

Light absorbers

The first requirement for light absorbers is efficiency, but they also need to show long-term stability in highly corrosive (aqueous) environments, says van de Krol. His team is particularly interested in using metal oxide-based semiconductors as light absorbers.

One of the materials they work on is bismuth vanadate, known by its chemical formula BiVO4 and also used as an inorganic yellow pigment. The material currently holds the efficiency record among metal oxide light absorbers, and is fairly cheap and easy to make.

The researchers are still experimenting with the material, modifying it by doping, nanostructuring, coating it with protection layers to enhance the stability, and so on. “We also modify the surface of BiVO4 with catalysts to enhance the performance,” says van de Krol. “One of the things we don’t understand is why BiVO4 works well with some catalysts, but not with others; we try to understand why, and use this knowledge to come up with new semiconductor-catalyst combinations.”

BiVO4 may be the best material right now, but it’s still not good enough; its solar-to-fuel efficiencies does not go beyond 10 per cent. “We need to find even better materials,” says van de Krol.

Quest for catalysts

Besides light absorbers, scientists are also searching for the perfect catalyst. Over the years, teams from around the world have developed prototypes of catalysts that can help split water into hydrogen and oxygen. But even though water splitting has been mastered in principle, many catalysts either require a lot of energy to get going, require either very high pressure and temperature or some expensive noble metals, or are very inefficient, says biochemist Petra Fromme of Arizona State University. Even worse, they are usually either very fast but not stable and disintegrate quickly, or relatively stable but very slow, she adds. “At the moment, the optimisation of molecular catalysts is like fishing in the dark.”

One major breakthrough came in 2011, when Daniel Nocera, a chemist now at Harvard University in Cambridge, MA, stunned the world with the first working ‘artificial leaf’. It looks nothing like a leaf, of course, and doesn’t convert water and CO2 into carbohydrates, as real leaves do. It is a rectangular piece of material the size of a postage stamp, a sandwich of inorganic materials with a photovoltaic wafer in the middle, which converts sunlight into wireless electricity. That electricity is routed to the outer layer of the ‘leaf,’ which is coated with chemical catalysts: one catalyst prompts the formation of hydrogen gas, the other oxygen. The hydrogen could then be stored, used as liquid fuel for hydrogen-powered cars, or converted to electricity with a fuel cell.

Nocera made his leaf using mostly cheap and widely available materials, and showed that it had about ten times greater efficiency than natural photosynthesis as well as much better stability over previous catalysts, able to work continuously in a lab for at least forty-five hours without a drop in activity.

It is Nocera’s cobalt phosphate catalyst, used as coating on one side of the leaf, that van de Krol’s team has turned to in order to enhance the performance of their light absorber. “We fully agree that cobalt is a promising catalyst - it works well, is not too expensive, and it can be scaled up,” says van de Krol.

Where the leaf falls down.

However, despite being promising, Nocera’s artificial leaf failed to become a marketable device. The main challenge for the concept, says van de Krol, is ion transport; ions have to diffuse from one side of the ‘leaf’ to the other side, over macroscopic distances of several centimetres, to close the electric circuit loop. “This leads to resistive losses that are very difficult to avoid,” he says.

To overcome this issue, he suggests having two electrodes that face each other. That, however, would make the design more complex. “In the end, it’s a trade-off between efficiency and simplicity, and more simple means lower cost,” says van de Krol.

While Nocera favours cobalt-based catalysts, researchers are looking at several other earth-abundant materials that could work as catalysts. Germany, for example, has just launched a large academic and industrial research programme to develop catalysts based on manganese. Scientists are also experimenting with nickel, iron and molybdenum.

“Nickel is widely used for the production of stainless steel, metal alloys and the current collector for batteries, and iron is nearly everywhere in our daily life, so both are very cheap,” says Michael Grätzel, a chemist at the École Polytechnique Fédérale de Lausanne.

Among the best candidates for water oxidation under alkaline conditions are Fe-Ni-O catalysts, while cobalt phosphide and molybdenum disulfide seem promising for hydrogen evolution under acidic conditions. “What’s really missing at the moment is an efficient earth-abundant catalyst for oxygen evolution under acidic conditions; here, we still do not have any alternatives for expensive catalysts based on, for example, ruthenium oxide and iridium oxide,” says van de Krol.


One of the main goals in the quest for a perfect light absorber and a perfect catalyst is efficiency, and that’s another problem faced by Nocera’s artificial leaf. He based his concept on a pH-neutral solution, for safety reasons: having large volumes of water with either very acidic or alkaline solutions would require extra safety precautions. However, in pH-neutral solutions the concentration of ions is very low, which means it has a sharply reduced efficiency, says van de Krol.

“We may be able to get around this by smart engineering solutions, but it will be challenging,” he adds. That’s why many researchers lean instead towards acidic or alkaline solutions.

In milestone after milestone, scientists are getting there. For example, a team at the Joint Center for Artificial Photosynthesis (JCAP) recently developed a device that converts 10 per cent of the solar energy it receives into fuel. This is much better than natural photosynthesis - after all, plants are able to convert only 1-2 per cent of sunlight into sugars and other carbohydrates.

The researchers also managed to achieve much better stability than before, by using a material called titanium dioxide to protect the photovoltaic materials from the corrosion that normally happens during the production of oxygen. The protective material is compatible with cheap catalysts that are used commercially.

While promising, the system is still not good enough. Despite achieving greater stability, for artificial photosynthesis-based devices to be mass-produced and to compete with other sources of fuel, a much higher efficiency than 10 per cent is needed.

Tackling climate change

While some researchers are mainly concerned with splitting water splitting to produce hydrogen fuel, others are working to put CO2 to use as well.

Researchers at the Lawrence Berkeley National Laboratory in August presented a system that uses water and solar energy to produce hydrogen, which in turn is used to produce methane, from CO2.

“Our artificial materials system has the same function as the green leaf: taking in CO2 and sunlight, turning it into acetate or methane - in our current case - and releasing oxygen,” says the group’s leader, chemist Peidong Yang.

The researchers developed a membrane arrangement of nanowires made of indium phosphide photocathodes and titanium dioxide photoanodes, able to harvest sunlight to split water into oxygen and hydrogen. The hydrogen was then transported to microbes that used it to convert carbon dioxide into methane.

“If you want to produce transportation fuel with high energy density, for example gasoline, you will have to start with CO2. And this is also an active approach for CO2 mitigation,” says Yang.

Remaining hurdles

Despite the rapid development of the field in the past decade, there are still many challenges to address before the technology will reach mass production. Hurdles range from improving the materials science to the engineering challenge of bringing all the components together in a workable device for artificial photosynthesis.

While scientists there are plenty of good candidates for catalysts, the choice of efficient light absorbers is still limited.

With costs high, it’s also tricky to scale up the existing systems. “The goal is to produce solar hydrogen at a cost of €5 per kilogram,” says Grätzel. “As global warming becomes more severe, CO2 emission taxes will make the solar fuel more economical. This will entice entrepreneurs to invest in scale up and production. Policy makers in the governments need to prepare solar fuel infrastructure.”

Meanwhile, he and many of his colleagues continue to be optimistic. “I believe that this technology will be implemented within the next 10 years,” says Grätzel. *

Recent articles

Info Message

Our sites use cookies to support some functionality, and to collect anonymous user data.

Learn more about IET cookies and how to control them