Electricity is an amazingly useful form of energy. Even better, we know how to generate it without carbon emissions. So step one in the battle to limit climate change is to shift as many energy services as possible over to electricity.
However, some uses of energy are not easy to electrify in the short term, maybe even in the long term. They rely on energy-dense liquid fuels.
So if we're ever going to get to net zero emissions, we badly need zero-carbon liquid fuels that can substitute directly for fossil fuels in those areas that can't be electrified.
The unavoidable (for now) need for liquid fuels
So what kind of energy services need liquid fuels? Let's focus on the US.
First and mostly, there's transportation. About 26 percent of US emissions come from transportation (90 percent of which is powered by fossil fuels).
Some of that can be electrified; around 60 percent of transportation emissions come from light-duty vehicles, which are already on the road to being electrified. Most of the rest, though, come from medium- and heavy-duty trucks and airplanes, which are difficult to electrify without serious advances in the energy density of batteries.
Then there's industry (another 21 percent of US emissions). Lots of high-heat industrial processes, like making steel and coke, require fossil fuel combustion. And industry uses lots of fossil fuels as feedstocks for chemical processes that make things like plastics. In both cases, there are no ready-to-hand substitutes for liquid fossil fuels.
What about biofuels? Well, they have proven something of a disappointment; the common ones compete with food crops, and the ones that use alternate, non-food crops never seem to bust out of the demonstration-plant phase.
So what to do? Happily, there is an alternative: solar fuels, sometimes known as artificial photosynthesis.
Plants, you see, absorb light, water, and carbon dioxide and produce oxygen and plant fuel. What if we could replicate that process, only make it faster and more efficient, and produce fuels that serve in place of fossil fuels?
Turns out it's possible! It's something scientists have been working on fitfully ever since the 1970s, but in the last decade or so the field has moved forward by leaps and bounds, with a range of breakthroughs. It's all still lab-bound at this point, but commercial viability is, if not right around the corner, at least on the horizon.
And if it pans out — if it becomes a viable alternative to liquid fossil fuels — then we will finally have a clear path to total decarbonization.
The two solar fuels: hydrogen, and stuff you make with hydrogen
There are many varieties of solar fuels being researched and tested right now. If you want the nitty gritty, see this 2015 report from MIT; this PNAS piece gets into detail about the latest research breakthroughs. I'll just run through a high-level summary.
The research falls into two basic buckets: The first produces hydrogen; the second uses hydrogen to produce hydrocarbon fuels.
1) Splitting water to extract hydrogen
Water, you will recall, is made of two atoms of hydrogen and one of oxygen (thus H2O). The energy from sunlight can be used to split them apart; the oxygen can be released and the hydrogen captured.
Hydrogen is a pretty nifty fuel in and of itself (though it's not nearly as energy-dense as fossil fuels like gasoline or methane). It can be stored in a fuel cell and later used to make electricity, compressed and transported in liquid form, or used as a feedstock in all sorts of chemical processes.
There are lots of ways to split water. Here's a table of the technologies being investigated:
Don't worry, there won't be a quiz!
The main thing to know is that alkaline electrolysis — where solar generates electricity and the current is passed between two electrodes that split the water — is not the most efficient method, but it is the best understood and best tested, and draws on PV infrastructure that's already built out. It will probably be the first to reach commercial application.
Photo-electrochemical processes eliminate the intermediate step of generating electricity. Light photons strike the photoelectrode directly, knocking lose electrons and holes, which in turn drive oxidation and reduction. This method is potentially more efficient, since it eliminates the conversions to and from electricity, but it still faces lots of lab challenges — mainly in finding the right materials. (It all takes place submerged in water, so corrosion is a problem.)
The "thermo" processes on the chart have to do with concentrated solar power, which uses mirrors to produce heat, which can then drive various chemical reactions. These approaches are also extremely promising, but also face many lab challenges before they're ready for commercial use.
So that's all about producing hydrogen. But hydrogen has its drawbacks. Storing it in a fuel cell or compressing it into liquid takes a lot of energy and reduces end-to-end efficiency. And using it on a large-scale basis would require new infrastructure.
2) Using hydrogen and carbon dioxide to produce hydrocarbon fuels
Which brings us to the second kind of solar fuel, the kind that reacts to the (newly freed) hydrogen with carbon dioxide to produce more energy-dense liquid fuels like methane (synthetic natural gas).
The carbon in carbon dioxide is split from the oxygen in one of the same three basic ways that the hydrogen in water is split from oxygen: electrolysis, photoelectrochemical methods, and thermochemical methods. It's much, much trickier, though.
"When compared to water splitting … conversion rates, efficiencies, and selectivity are low," the MIT report notes. "Considerable work is needed to identify more efficient processes and catalysts for carbon dioxide reduction by these more direct methods."
There's also the additional problem of where to find a steady supply of carbon dioxide to use as a feedstock. Carbon-capture methods at power plants and industrial facilities remain extremely expensive. And then what happens when fossil fuel power plants shut down — where does the carbon dioxide come from?
The reason these problems are worth tackling is that energy-dense fossil fuel substitutes like methane could, in combination with clean electricity, get us all the way to zero emissions.
Relative to hydrogen, methane has myriad benefits: It is much easier to transport, integrates more easily with existing pipeline infrastructure, substitutes more easily for fossil fuels, and can be used as a feedstock for fertilizer and plastics. Synthetic natural gas could potentially do everything natural gas does today, with none of the emissions. And other solar fuels like methanol or ethanol could be burned in today's gas engines, without any retrofitting.
There's a bit of poetry to all this. It took nature millions of years to transform sun, water, and carbon dioxide into energy-dense fossil fuels, via photosynthesis and high-pressure compression. Now humans have figured out a way to re-run the process in a lab, millions of times quicker, with none of the carbon. Clever, clever humans.
Here's a graphic from MIT that tries to get it all in one place:
No commercial products yet, but lab breakthroughs aplenty
So where does all this stand now? For a great overview, see that PNAS article, which covers three recent lab breakthroughs.
The first, by Leone Spiccia at Monash University in Victoria, Australia, is in basic electrolysis.
Spiccia used high-performance triple-junction solar cells to generate electricity. The electricity passes through nickel-foam electrodes to catalyze water splitting. The system converts solar energy into hydrogen fuel with an efficiency of 22%. Spiccia is now working on reducing inefficiencies in the connections between the parts, and he believes that an overall efficiency of 28% or 30% is possible.
The hydrogen thus produced is still more expensive than what's produced the more conventional way (steam reforming methane, which is highly energy intensive and polluting), but it's getting closer and closer.
The second is from the Joint Center for Artificial Photosynthesis (JCAP), at Caltech, which has developed a version of photoelectrolysis, which eliminates the need for a separate solar cell by eliminating the need for electricity.
As part of JCAP, [Caltech chemist Nathan Lewis] developed a water-splitting system with electrodes that are something like submerged photovoltaic panels. His system looks like a sealed reactor full of water, illuminated from the outside, shiny photodiodes within. As in an ordinary solar cell, light strikes a semiconductor, generating electrons and positively charged "holes." But rather than funnel these off to an electrical grid or a battery, the JCAP device passes them directly to catalysts to split water.
This direct route is more efficient, without the conversion to and from electricity. The challenge is finding electrodes that don't corrode underwater and protecting them with materials that reduce their efficiency as little as possible.
The third is by Peidong Yang at UC Berkeley. Yang saw that chemists were having a hell of a time finding synthetic catalysts that could effectively react hydrogen and carbon dioxide to create hydrocarbon fuels.
That’s because the chemistry is much more complex. Splitting a molecule of water takes four electrons, says [Jens Norskov, a professor at Stanford]. Making the simple hydrocarbon methane is a reaction involving eight electrons, each with different energies, which have to be shuffled around through several steps to create the single-carbon molecule.
Finding synthetic catalysts that will do that work — without degrading, or healing and regenerating themselves if they do degrade — is a bear.
But nature does it all the time, easy as pie. So Yang is putting nature to work. To make a long story short, he's coating one of the electrodes with bacteria. (There are also nanowires involved; don't ask.) The bacteria absorbs hydrogen gas, combines it with carbon dioxide, and poops out methane.
There are other labs investigating various other "living catalysts." Just last week, Daniel Nocera (a long-time solar fuels pioneer) and Pamela Silver at Harvard published a study in Science demonstrating artificial photosynthesis at 10 percent efficiency (well above nature's benchmark). They used hydrogen to grow a microbe which then absorbs hydrogen and carbon dioxide and poops out usable fuels — "in principle," Silver said, "any downstream carbon-based molecule." So far they've made isopropanol, isobutanol, isopentanol, and PHB, a bio-plastic precursor.
As Yang concedes, synthetic catalysts would be preferable (living ones are finicky — they care about temperature and pH), but as yet, humans aren't quite clever enough to replicate what nature does. So why not let nature do it?
Despite all these impressive breakthroughs, serious commercial application of solar fuels remains a ways off. Much research remains to be done, not only to find cheap and efficient ways of producing hydrogen, and the best materials for catalysts, but also to establish reasonably cheap carbon capture.
This will require concerted research and development efforts in a number of key areas including photovoltaics, electrolysis and fuel cells, catalysts, efficient CO2 collection, hydrogen storage and distribution, and synthetic fuel production from CO and H2 feedstocks. Only a joint and concerted effort by government, industry and academia will lead to measurable progress in this critical endeavor.
That last sentence cannot be emphasized enough.
When combined with zero-carbon electricity, carbon-neutral liquid fuels finally illuminate a path to truly carbon-free energy and agricultural systems. A sane country — a sane world — would be pouring billions of dollars into accelerating their research and commercial deployment.