How to Make Hydrogen

One of the more frustrating phrases from BEV fanatics is “99% of hydrogen is from fossil fuels, so hydrogen vehicles pollute a lot.” There is a lot wrong with the statement, but from a prior article we know that most H2 is used in industry, not vehicles, so conflating their supply chains doesn’t make sense. Here we will discuss the major ways H2 can be made: from fossil fuels, from electricity, from other less scalable means, and by pulling it from the ground.

The major scaled or scalable pathways to hydrogen production

Most of these pathways can mitigate their emissions by sourcing better inputs or capturing and sequestering their emissions. Here’s all the high level detail you need to know to better understand this:

Electrolytic H2

Electrolysis is splitting water into hydrogen and oxygen with electricity. Putting it back together in a fuel cell can create electricity, or the hydrogen can be used with conventional uses.

When connected only to renewables, electrolyzers make hydrogen with zero emissions. When connected only to the grid in most of the US, they produce about 2.25x the emissions of our current fossil H2 mechanisms. Note that this is currently how BEVs operate – they connect directly to the grid and have significant associated emissions.

Best case scenario, an electrolyzer is hooked up to renewables and delivers hydrogen to an already-existing network. In reality, electrolyzers have largely been connected to the power grid. In the coming years, expect electrolyzers to be connected to renewables with backup from the power grid.

The broad category of electrolysis can be subdivided several times. First we have major types (alkaline, SOFC - Solid Oxide Fuel Cell, PEM – Protonic Exchange Membrane, AEM – Anionic Exchange Membrane) and then we have various membrane types and membrane structures (dry anode vs other setups). Most investors won’t need to know the details themselves, but they’ll need to hire people to understand what the niches of each type are. I’ll go into more details of this in a future post. The shortest version is that alkaline is old tech, reliable but has issues, PEM is newer tech but expensive and has resource constraints, AEM is future tech, and SOFC is very high temp and has very specific use cases.

After electrolyzer costs come down, electrolytic H2 will be the lowest cost way of producing H2 in parts of the world with abundant land and a lack of natural gas.

Fossil-derived H2

Using fossil fuels or coal to make hydrogen is a process that strips hydrogen from hydrocarbons, leaving the carbon behind. Fossil-derived H2 is primarily from natural gas and coal.

Natural Gas Reformation

This reaction converts natural gas to hydrogen and carbon (CH4 + O2à CO2 + 2H2). The amount of total emissions depends on feedstock and whether there is some sort of carbon capture at the back-end. Here are details of both:

Upstream GHG emissions controls

It turns out that natural gas has all sorts of different ways of sourcing it. From least emissions to most, we have:

1. Renewable natural gas: This is natural gas made from biological or other feedstocks. The general idea is that it is fugitive biological methane or other gases with high greenhouse warming potential are trapped, captured and sold, so it reduces greenhouse warming.

This methane is more expensive than naturally occurring methane, but subsidy regimes like Low Carbon Fuel Standards and now 45V make the price break-even or better than methane. It has two main issues. First, there is not enough RNG production to replace our natural gas use – this is a limited scale pathway. Second, a lot of biomethane is low quality and will break equipment if used directly.

2. Low emission, certified, or responsibly sourced natural gas: Current methane production systems leak quite a bit during initial drilling, production, and transport. Methane is a very strong greenhouse gas, so these upstream leaks are a significant contributor to the GHG emissions of any natural gas system. Closing off these leaks and having it consistently verified will reduce the associated upstream emissions and reduce total emissions

Downstream GHG emissions controls

To capture and store the carbon can be quite intensive, however. Some types of hydrogen production have a pure stream of CO2 on the back end and can be captured and stored fairly cost-effectively. Older types have a dilute carbon stream that is mixed with oxygen, nitrogen, and other pollutants and they require extensive and complicated chemistry to capture and store the carbon.

Major types of reformers

1. Steam Methane Reformers – Also known as SMRs (not to be confused with small modular reactors. These take in natural gas, combine it with atmospheric air to make syngas, a mixture of carbon monoxide and hydrogen, and then combine it with steam to make CO2 and H2. It then separates these using a special filters

2. Auto-thermal reformers (ATRs) – the next generation of steam methane reforming, originally made during the Apartheid in South Africa to create higher end fuels from natural gas or coal. The primary difference here is that they use a stream of pure oxygen for the initial reaction, so the final product is pretty much pure hydrogen and CO2, making the CO2 much easier to sequester. ATRs are more costly up front but more efficient and far easier to capture CO2 from than SMRs

Coal Reformation

Coal has a lot of carbon and a lot less hydrogen compared to natural gas. As a result, when coal is gasified to make hydrogen, it produces ~2.5x the CO2 compared to using natural gas. Worse, a coal gasifier is massively expensive compared to an SMR or ATR, so the economics aren’t great anywhere with abundant natural gas. Coal gasifiers also produce prodigious amounts of other pollution. So where are these used? Specifically in locations with small natural gas resources, large cheap coal resources, and a general disregard for public health vis-à-vis air pollution. We’re talking about China and India specifically.

There were a lot of trials to make coal gasification work in the US, but the coal gasifiers are just too expensive and the carbon capture would add a massive additional cost. Coal gasification is not a future technology and will die out as soon as electrolyzer costs come down. Nonetheless, it’s currently in the area of a double-digit percentage of the hydrogen production outside the US.

Pyrolysis and other oxidation techniques

With only a few exceptions, these are not hydrogen production mechanisms. The startups doing these will tell you differently, but most of these technologies were meant to create a substance called Carbon Black which is used in tires, cosmetics, and a few other things, and H2 is a byproduct.

One example is a company that produces carbon black with hydrogen as a byproduct using plant matter. In the DOE I did a calculation of their H2 costs without carbon black subsidizing the H2 cost, and it came out to $18/kg. You’ll see that a significant number of these systems are not real hydrogen production, but instead have another purpose and they are jumping on the H2 hype train.

Other techniques like partial oxidation and some newer pyrolysis systems are designed from the ground up to be for hydrogen production. Know the difference before you put your money down or before you partner with one of these techs.

Other production types

Most of these aren’t scalable, ignore systems costs, or ignore other issues. These are your “I turn waste biomass / waste water / trash into hydrogen.” For the most part, these systems have an input problem: the material coming in is so variable that the systems break a lot and the output isn’t consistent enough. They also struggle to source sufficient material to scale. Usually they are solutions to different problems and are looking to jump onto the H2 hype train or to get subsidies for H2.

Generally these rely on “tipping fees” - payments to take waste and convert it into something useful. Major pathways you’ll see are trash (with tipping fees), dead trees (policy measure to deal with dead trees), other plant matter, or wastewater. Most of these will always require some sort of strong policy incentive to be viable.

Geologic Hydrogen

This is either fairy dust and will never come to fruition, or if it does come to fruition is will be a silver bullet to help address climate change. Geologic H2 is similar to fossil fuels in some ways, but it’s still clean. Fossil fuels are largely decayed organic matter than has been subject to heat and pressure underground. Geologic hydrogen is rock and water that has been subject to heat and pressure underground. Both of them get trapped underground and come up through a drilled well, but we barely know anything about geologic hydrogen.

The amazing part about geologic hydrogen is that it could be clean and very inexpensive. Right now it’s $1/kg to make hydrogen from natural gas in the US. Geologic hydrogen could be zero emission and cost between $0.50 to $1/kg. If it’s real and large-scale, it will be an absolute game changer.

Summary

If you’re an investor or you’re looking at vendors, this area is very fraught ground. Bring on a consultant with actual hydrogen experience (IE not the big three strategy ones) to help you through this.

The type and pathway of hydrogen determines a lot of attributes and commercial viability. Most importantly, it determines whether a company or tech is hype or whether it is investible. Ultimately, the feedstocks and production methods determine how clean it can be as well. In closing, here’s a helpful table: