Cleaning Up Hydrogen
According to the Nature Conservancy, the average American contributes 16 tons of carbon to the atmosphere per year, which we need to reduce to 2 tons/year by 2050 in order to avoid a 2 degree Celsius increase in global temperature. How do we do that?
On a personal level, you can go solar and electrify everything. But, there are some things you can’t control, like the emissions from that cross-country flight for work or from manufacturing the steel and cement that support your office, your favorite ballpark, and your grocery store. These industries are tougher to electrify for various reasons, including battery weight and intermittancy. If these industries are going to decarbonize, hydrogen might be a big part of the solution.
The three largest U.S. hydrogen producers are Air Products, Linde, and Air Liquide. Much of the rest is distributed among large oil and gas companies. Several production facilities are concentrated along the Gulf Coast, but other sites are spread across the country.
As the world seeks to decarbonize, hydrogen will be called upon to serve some difficult industries. According to the IEA, global hydrogen production will need to grow over 50% by 2030 in order to meet the net zero scenarios, from about 95 Mt produced per year to 150 Mt/year.
However, as of today most (over 99% of) hydrogen production itself is not clean. Traditional “gray” hydrogen is produced through fossil gas (methane), which produces not only an impure hydrogen (H2) but also carbon dioxide (CO2). According to the EPA’s GreenHouse Gas Reporting Program, over 41 million metric tons of carbon were emitted by U.S. facilities producing hydrogen in 2021. That’s equivalent to the average emissions of 2.5 million Americans—the population of Chicago—from the production of hydrogen alone!
Green and Blue Hydrogen
Fortunately, new and cleaner methods of hydrogen production have arrived. The two most prominent are “blue” hydrogen and “green” hydrogen. Blue hydrogen is produced the same way as gray hydrogen but with carbon capture, utilization, and storage (CCUS). Depending on the CCUS efficiency, some or most of the carbon pollution can be avoided.
There is another way to produce hydrogen called electrolysis, which relies on electricity to split water (H2O) into hydrogen (H2) and oxygen (O2). The good news is that it releases only hydrogen and oxygen, no carbon; the bad news is that it requires a tremendous amount of electricity, and producing electricity often involves carbon emissions.
“Green” hydrogen requires that the hydrogen-producing electricity comes from fully renewable sources, such as solar or wind. (A variation called “pink” hydrogen is also low-emissions but not renewable, produced through electrolysis powered by nuclear energy.) As of now, blue and green hydrogen account for less than 1% of the total hydrogen production.
The major reason for gray hydrogen’s dominance is its cost effectiveness, averaging around at $2/kg. The Department of Energy’s Hydrogen Shot aims to reduce the cost of clean hydrogen (blue and green) from $5/kg to $2/kg by 2025 and $1/kg by 2030, making it more than competitive on cost against legacy methods of production.
This producted cost reduction is being driven by economies of scale, gained through significant investment from major gas/chemical companies as well as the IRA. Companies like Air Products, Linde, and Air Liquide have announced billions of dollars of investment in new blue and green hydrogen facilities in the U.S. and internationally to expedite their role in cleaner hydrogen production and the broader energy transition.
Shades of Blue and Green
So, we just put in a lot of CCUS, electrolyzers, and solar panels and we’re on our way? Unfortunately, it’s not quite that simple. Not all blue and green hydrogen is created equal.
As alluded above, different methods of blue hydrogen CCUS can result in different levels of emission reductions as described by the International Energy Agency. While gray hydrogen emissions range from 10-14 kg CO2 / kg H2, common CCUS methods reduce emissions to 5-8 kg CO2 / kg H2 (~50% reduction) and advanced methods can achieve 0.8-6.0 kg CO2 / kg H2 (up to 90% reduction) though are not yet in wide operation. IRA credits kick in below 4 kg CO2 / kg H2 and increase dramatically below 0.45 kg CO2 / kg H2, incentivizing further improvements in CCUS.
Green hydrogen has more IRA incentives in its favor, but its carbon accounting can be even more challenging. Depending on how you measure the carbon intensity of the electricity produced, the lifecycle greenhouse gas emissions from electrolysis-produced hydrogen will vary wildly.
Let’s see what happens when you power your electrolyzer from grid electricity.
Carbon Intensity of Electrolysis
As best I can tell, S&P Global’s location data for proposed blue and green hydrogen facilities is not available without a (very expensive) subscription, so I’ve partially recreated it through my own effort. I filled in the latitude and longitude data on this spreadsheet taken from the EPA’s Flight database. It is not a complete list; time-permitting, I may go back and fill in the complete list a later date. I have started with the largest proposed facilities and hope that this partial list proves to be a sufficient demonstration.
If we look at the green hydrogen sites specifically, where electrolysis will be used to produce hydrogen, grid emissions vary significantly, from CAISO’s annual average of 225 g/kWh to MISO’s 566 g/kWh. (I sourced the grid regions from Electricity Maps and added in the annual emissions data by balancing authority zone.) Multiplying the year’s average carbon intensity by the estimated kWh needed for full production at the site (assuming ~50 kWh per kg of H2), we can see which proposed site would generate the most Scope 2 emissions: Hydrogen City near Corpus Christi, TX.
Fortunately, these sites are planning to produce green hydrogen, meaning they will not be using grid electricity but procuring their own. Assuming they all use solar (a life cycle assessment of 50 g CO2 / kWh), my estimates suggest that their CO2 reductions range from 78% to 91% compared to the use of grid electricity.
Furthermore, the estimated ratio of CO2 produced per unit of H2 from grid electricity falls in the 21-28 range (aside from the relatively clean grids of NY and CA), consistent with other estimates. Meanwhile, the estimated ratio of CO2 to H2 from solar’s life cycle assessment is 2.5, and estimates of LCA emissions from wind turbines are even lower. Comparitively, gray hydrogen is found to have a CO2:H2 ratio in the 10-15 range, and blue hydrogen is about 7. This analysis supports the idea that green hydrogen (powered by wind or solar) is the leading option for reducing carbon emissions for hydrogen producers.
Further Debate Over Green Hydrogen Rules
The simplest way to ensure that fully renewable energy is used in hydrogen production is to establish an off-grid facility, meaning the electricity is produced entirely on-site and not taken from the local grid and mixed with other sources. However, in order for an electrolyzer to optimized, it needs to be running as often as possible, not just when the sun is shining on solar panels or the wind is powering turbines. As the EU Rules for Renewable Hydrogen note, “While off-grid is the simplest approach to ensuring that the electricity used to produce hydrogen is 100% renewable, it limits operation to the periods when renewable electricity can be produced, or requires additional investment into electricity storage.”
This added hassle and expense prompts many hydrogen manufacturers to consider on-grid solutions with renewable offsets. Even though the offsets are for fully renewable energy, exactly how, when, and where those offsets are procured effects the effective carbon intensity of hydrogen production. The European Union rules frames these criteria as additionality, temporal correlation, and geographic correlation. I’ll save that discussion for a future post.
In the meantime, Congress (via the IRA) left these details to the Treasury Department, which has already missed the one-year deadline to release their definitions allowing the debate over the guidelines to persist. Within the next month, we may find out how the government will handle these factors, as well as renewable natural gas credits, when handing out a decade’s worth of credits for the production of hydrogen.