Hydrogen for Decarbonization: A Realistic Assessment

Introduction

Hydrogen is increasingly seen as an option for decarbonizing certain emissions sources and sectors where direct electrification or other low-carbon options might not be technically or economically feasible. Interest in hydrogen is not new and can be traced back to the 1970s, but previous attempts to develop this potential energy carrier – most recently in the early 2000s – never succeeded in establishing hydrogen as a significant alternative to conventional fossil fuels. Some argue that this time is different, and in some important regards, it is. Today’s enthusiasm for hydrogen is primarily driven by climate concerns, rather than by energy security considerations. In addition, the unprecedented scaleup of renewable energy production and a rapid decline in costs for wind and solar power have given new life to the concept of producing hydrogen from renewable electricity.

What has not changed, however, are the physical and chemical properties of hydrogen itself, including, notably, its low volumetric energy density, which make hydrogen difficult to store and transport. This and other properties have historically hindered hydrogen’s entrée into new applications beyond the industrial processes it has traditionally been used for.

Nonetheless, optimism about hydrogen’s potential as a tool for decarbonization continues to drive techno-economic assessments that lack grounding in facts and experience. This has significant ramifications, starting with a tendency to generate unrealistic estimates of the likely cost and future supply of clean hydrogen.

In addition, many current assessments minimize the challenge of utilizing hydrogen in various end-use applications and fail to recognize tradeoffs in terms of supporting other technologies that are needed to enable economy-wide decarbonization.

This short paper presents a pragmatic framework in which hydrogen can play an effective role as a climate solution while acknowledging the decarbonization challenge posed by today’s hydrogen production. Viewed through this lens, today’s engineers, planners, and policy makers can better identify high-potential and cost-effective applications for clean hydrogen.

Hydrogen Basics

Hydrogen Production Today

Virtually all dedicated hydrogen production today uses fossil-fuel feedstocks. Globally, about 59 million metric tons or ‘tonnes’ (MT) of hydrogen are produced annually from natural gas, using steam methane reforming. Another 20 MT per year (MT/y) are produced from coal, using gasification, with the balance of global production from oil and electricity (most coal-based production is located in China). These methods are carbon intensive. About 16% of hydrogen production in 2022 consisted of by-product hydrogen, which is most often consumed on site.

Figure 1: Hydrogen Production by Technology, 2020 – 2022

Source: International Energy Agency 2023 (Adapted by CATF)

Methods of Decarbonizing Hydrogen Generation

Direct carbon dioxide (CO₂) emissions from current modes of hydrogen production are estimated to account for approximately 2% of global CO₂ emissions – including about 240–380 MT/y from refining and approximately 680 MT/y from industry (IEA 2023). Thus, existing hydrogen consumption provides an appealing opportunity for decarbonization. To effectively decarbonize hydrogen production at scale, two main pathways – water electrolysis and methane reforming with carbon capture and storage (CCS) – are worth focusing on.

In water electrolysis, an electrical current is used to split water into its constituent elements. So long as the electricity is generated in a low-carbon manner, the resulting hydrogen can be considered low-carbon (to qualify as ‘clean’ or ’low-carbon‘ for regulatory or policy purposes, the electricity source may need to meet specific requirements or a specified threshold of carbon intensity). Water electrolysis is significantly more expensive than conventional, carbon-intensive methods of hydrogen production, and currently accounts for a tiny fraction of global hydrogen supply.

Methane reforming with carbon capture and storage uses natural gas as a feedstock. To be low-carbon, this production method must include CCS. Lifecycle carbon emissions from this hydrogen production pathway include upstream methane emissions from natural gas production and transmission, indirect CO₂ emissions from electricity consumption, and direct CO₂ emissions from the reforming unit itself. To qualify as low-carbon, these emissions must be minimized to the maximum extent practicable, through high rates of carbon capture, low methane leakage, and use of low-carbon electricity. For example, to qualify as low-carbon under European regulations, these direct emissions must be abated by 90% or more using carbon capture technology.¹

Supply of Clean Hydrogen

Despite a plethora of recent project announcements, low-carbon hydrogen is currently in short supply. In 2022, water electrolysis accounted for only 0.1% of global hydrogen production, mainly due to the challenging economics of this production pathway but also due to an undeveloped supply chain. While supportive financial incentives and policies are set to change this picture in the near term, significant hurdles remain. Large-scale hydrogen production facilities are capital intensive and require abundant supplies of clean electricity (in the case of water electrolysis) or highly effective carbon capture systems with other supporting infrastructure such as CO₂ pipelines and geologic storage facilities (in the case of steam methane reforming with CCS).

Either type of hydrogen production facility is likely to require additional years for design, planning, permitting, and delivery. Furthermore, large-scale water electrolysis facilities will compete for clean electricity at a time when demand for renewable generation to decarbonize the power sector as a whole is high and growing. Moreover, direct grid decarbonization can be expected to take priority because of its significant near-term carbon abatement potential. To put these tensions in perspective, supplying all of current U.S. dedicated hydrogen production (10 MT/y) using water electrolysis would consume more than half (60%) of all current U.S. electricity generation from renewables.²

Global Hydrogen Trade

Virtually all hydrogen produced today is consumed at the site of production or transported by pipeline to a relatively nearby facility; only minor quantities are consumed at a distant location or transported by industrial gas companies in the merchant business.

Figure 2: Hydrogen transportation and storage

Table 1: Physical Properties of Hydrogen

Source: International Energy Agency 2019

Property	Hydrogen	Comparison
Density (gaseous)	0.089 kg/m3 (0˚C, 1 bar)	1/10 of natural gas
Density (liquid)	70.79 kg/m3 (-253˚C, 1 bar)	1/6 of natural gas
Boiling point	-252.76˚C, 1 bar	90˚C below LNG
Energy per unit of mass (LHV)	120.1 MJ/kg	3x that of gasoline
Energy density (ambient cond., LHV)	0.01 MJ/L	1/3 of natural gas
Specific energy (liquefied, LHV)	8.5 MJ/L	1/3 of LNG
Flame velocity	346 cm/s	8x methane
Ignition range	4-77% in air by volume	6x wider than methane
Autoignition temperature	585˚C	220˚C for gasoline
Ignition energy	0.02 MJ	1/10 of methane

Hydrogen’s physical properties (Table 1) highlight why it is a difficult molecule to store and transport: it has very low volumetric density (like helium, hydrogen exists as a gas at ambient temperature and pressure, but it is twice as light) and a boiling point that is a few degrees above absolute zero, the lowest temperature theoretically possible.

The challenges of transporting and storing hydrogen have not deterred numerous resource-rich countries from announcing bold plans to produce and export hydrogen on a large scale. Indeed, some of these countries have signed hydrogen export agreements with traditional energy importers. Similarly, some industry analysts predict that global trade in hydrogen will eventually mirror current trade in oil and natural gas – in both magnitude and reach.

As a basis for these high expectations, advocates point to a combination of increased policy-driven global demand for low-carbon hydrogen to help meet climate commitments and the cost advantage that certain resource-rich countries expect to have as producers of low-carbon hydrogen compared to energy-hungry but resource-poor importing nations. However, it is not clear that current projections adequately reflect the techno-economic realities of hydrogen transportation. These realities are explored in a 2022 analysis, commissioned by Clear Air Task Force (CATF), which analyzed various pathways for importing hydrogen to Europe and highlighted the limitations of maritime transport, which would be the only alternative to pipeline transport for moving large quantities of hydrogen across significant distances. Options for liquid hydrogen carriers that could be suited to transport by ship are described in the box.³

Figure 3: Hydrogen Production and Transportation Supply Chain

In summary, hydrogen’s intrinsic physical properties make it such that maritime transport of hydrogen is costly and/or inefficient. Unfortunately, these properties cannot be changed by innovation or technology. Currently, all developers of low-carbon hydrogen are producing and exporting products – such as ammonia, methanol, sustainable aviation fuels, and steel – that consume hydrogen as a feedstock. Significant volumes of hydrogen can theoretically be transported by pipeline, but geopolitics are likely to be a decisive factor in whether pipeline transport comes to play a major role, especially in the context of North African and European trade.

Prioritizing End Uses for Clean Hydrogen

Today, global hydrogen consumption totals around 95 MT/y – almost all for use as a feedstock (not a as a fuel) in refining⁶ (41 MT/y), ammonia production (33 MT/y), methanol production (16 MT/y), and steel manufacturing (5 MT/y). For clean hydrogen to play a major role in decarbonization, uses of hydrogen in fuel applications would have to expand dramatically.

Given the limitations discussed in previous sections with respect to physical properties and potential costs, low-carbon hydrogen is best used in applications where there are simply no other good decarbonization options. Ideally, market forces, shaped by increasingly stringent decarbonization policies, will direct low-carbon hydrogen to the optimal end-uses. However, industrial policy should encourage—or at least not discourage—economically and environmentally efficient outcomes. The next sections discuss an approach to prioritizing applications for clean hydrogen as policy makers and investors weigh difficult tradeoffs in achieving decarbonization objectives over the decades ahead.

Figure 4: Hydrogen Use in Industry

Source: International Energy Agency 2023 (Adapted by CATF)

Figure 5: Hydrogen Applications

No-Regrets Applications of Low-Carbon Hydrogen

Replacing existing carbon-intensive hydrogen consumption with low-carbon hydrogen is a no-regrets pathway to reduce carbon emissions, given established industrial uses of hydrogen and associated infrastructure. This approach can also ensure the sustainable continuity of production where no alternatives to hydrogen exist.

Crude oil refining – Refineries produce a wide array of products critical to the functioning of today’s economy, and hydrogen is a critical feedstock in their operations. Many of these products are hard to replace quickly and economically, and they are likely to remain in our future economy as well. Hydrogen is used, for example, to remove sulfur, nitrogen, oxygen, olefins, and heavy metals in transportation fuels. These hydrotreating operations are necessary to ensure that finished petroleum products satisfy technical, government, and safety requirements. Hydrogen also plays a role in increasing the product yields from hydrocracking operations and to produce a variety of non-fuel products such as lubricants and anode grade coke, a key component in the production of steel and aluminum. Using low-carbon hydrogen to replace 26 MT/y of unabated hydrogen production in refineries could result in 240–380 MT/y of CO₂ emission reductions (equivalent to UK emissions)– a ‘low-hanging’ decarbonization opportunity.⁷

Figure 6: Refinery Non-transportation Fuel Products

Ammonia production – Ammonia is a critical ingredient in nitrogen fertilizers, which play an essential role in providing a secure food supply for human populations worldwide. In fact, 70% of global ammonia supply goes to fertilizer production.⁸ Ammonia also has other important uses such as for explosives in the mining sector, synthetic fibers, and specialty applications. Current ammonia production – at 190 MT/y worldwide – is estimated to generate 450 MT/y of global CO₂ emissions⁹ (approximately 70% of ammonia is produced using natural gas; the balance is produced using coal, mainly in China). Hydrogen is an intermediate input in ammonia production, which involves reacting hydrogen with nitrogen from the atmosphere. Given the critical role of ammonia in underpinning the modern, industrial-scale agricultural system on which all of humanity depends, decarbonizing the hydrogen used to make ammonia should rank high on the list of applications for low-carbon hydrogen.

Methanol production – Methanol (CH₃OH)¹⁰ s a key industrial chemical used in the production of formaldehyde (an intermediate chemical that is used to produce polyurethane and various resins), acetic acid (an intermediate chemical that is used to produce adhesives, latex paints, various resins, and sealants), and plastics (which involve converting methanol to olefins). Methanol and its derivatives are also used as fuel additives to improve combustion properties. Current methanol production – at 98 MT/y worldwide – is estimated to contribute 130 MT/y of global CO₂ emissions (60% of methanol is produced using natural gas; the balance is produced using coal, mainly in China, which accounts for half of global methanol production).¹¹ Hydrogen is an intermediate input in methanol production. Given the importance of methanol for a variety of industrial applications, decarbonizing the hydrogen used to make methanol should rank high on the list of applications for low-carbon hydrogen.

Steel manufacturing – Hydrogen currently plays a role in steel manufacturing via the direct reduced iron-electric arc furnace (DRI-EAF) process (known by the trademark Midrex®), in which hydrogen from a synthetic gas (mainly H2+CO) is used to remove oxygen from DR-grade iron ore. The idea of using low-carbon hydrogen in existing DRI applications has been proposed as a way to reduce climate impacts from steel manufacturing. Interestingly, the Middle East and North Africa (MENA) region shows great decarbonization potential in this regard: although this region accounts for just 3% of global crude steel production, it accounts for 46% (55 MT/y) of the world’s total DRI-based steel production.

Other Potential Applications of Low-Carbon Hydrogen

Fully decarbonizing transportation entails complicated challenges, particularly for those forms of transportation that are difficult to electrify. Hydrogen will be required to produce fuels for these segments of the transport sector.

Aviation – Decarbonizing the aviation sector will require low-carbon hydrogen to upgrade biomass-based sustainable aviation fuels (bio-SAF), to synthesize jet fuel from hydrogen and captured carbon (synthetic SAF), and, potentially, to power aircraft that directly utilize hydrogen fuel. SAFs draw interest because they have the advantage of compatibility with existing infrastructure and engines (for this reason, they are often called ‘drop-in’ fuels). Oils and fats of biogenic origin to produce bio-SAF must be treated with hydrogen to produce straight chain paraffinic hydrocarbons with no aromatics, oxygen, or sulfur.¹² Using low-carbon hydrogen in the production of biomass-based transportation fuels could help reduce associated lifecycle emissions. However, as highlighted in a recent CATF report, land-use and supply chain constraints on biomass feedstocks mean that other fuel options will need to be developed, including synthetic fuels (or ‘e-fuels’) produced using a combination of hydrogen, electricity, and CO₂ sourced from non-biogenic feedstocks.

Synthetic fuel production, however, is currently technically and economically challenging. If all flights between JFK and Heathrow airports were to run on e-fuels, for example, a facility the size of the NEOM Green Hydrogen Complex would be required just to supply the quantities of hydrogen needed to produce these fuels.¹³

Heavy-duty trucking – A recent CATF analysis shows that hydrogen fuel cell electric vehicles (FCEVs) can play an important role alongside battery electric vehicles (BEVs) in decarbonizing the trucking sector. Relative to diesel trucks, heavy-duty FCEVs can complete long-haul routes without a substantial number of additional refueling stops, can be refueled in approximately the same amount of time, and their powertrains are only slightly heavier – such that FCEVs can carry nearly all the cargo that diesel trucks carry when fully loaded. To the extent that heavy-duty FCEVs are a strong replacement candidate for diesel on long-haul routes, this would increase the share of the overall truck fleet that can be decarbonized since FCEVs would offer the same operational flexibility as diesel vehicles and avoid the time penalties associated with charging BEVs. The role of FCEVs, however, will also be influenced by other factors such as total cost of ownership and well-to-wheels lifecycle emissions, which can exceed those of a fully electrified pathway (i.e., BEVs). For instance, an FCEV running on low-carbon hydrogen made from electrolysis using low-carbon electricity can achieve similar lifecycle emissions to a BEV, but it will require approximately 2.5 times the electricity of a BEV to travel an equivalent distance. Hydrogen can be produced off-site, but the economics of transporting hydrogen by truck to dispersed fueling stations presents an additional challenge. Alternatively, sourcing low-carbon hydrogen from natural gas reforming with CCS and strict upstream methane emissions control can alleviate the loads on the electric grid, but this pathway may achieve slightly lower emissions reductions. Given the centralized nature of natural gas reforming plants, it could also increase transportation demands for getting hydrogen to refueling stations.

Marine shipping – Low-carbon ammonia is a strong contender as an alternative marine fuel. However, health, safety, and environmental concerns associated with bunkering, storing, and combusting ammonia in marine engine rooms need to be thoroughly examined at a systems level before large-scale use of ammonia can be cleared for a potential role in decarbonizing marine transport on the open seas. Another alternative marine fuel capable of reducing lifecycle carbon emissions is methanol – indeed, numerous modern cargo ships
are being built with dual fuel capability to handle marine diesel oil today and methanol in the future. Unlike ammonia, methanol emits CO₂ when it is burned, which means that reducing lifecycle emissions requires sourcing ‘sustainable’ carbon atoms for the methanol production process.

Niche Applications for Low-Carbon Hydrogen

Power generation – The appeal of using low-carbon hydrogen in power generation is easy to understand: it can replace natural gas but emits no CO₂ when burned. However, this application entails numerous technological, infrastructure, and system challenges beyond simply operating turbines on hydrogen. For example, the quantities of fuel needed for power generation would likely necessitate geologic storage of hydrogen and dedicated transmission and distribution pipelines. Associated costs make it important to focus on the carbon intensity of any hydrogen used. Two options – hydrogen from natural gas with CCS and strict upstream methane emissions control or from electrolysis with clean electricity – are discussed briefly below.

For hydrogen produced from natural gas with CCS, a carbon intensity of 3 kg CO₂-equivalent per kg of hydrogen is on the lower end of what is technically achievable. Combusting this ‘blue’ hydrogen in a simple-cycle power plant reduces lifecycle emissions from the power plant by roughly half compared to natural gas combustion. This level of emissions reductions, combined with the cost of producing hydrogen using natural gas with CCS, results in CO₂ abatement costs significantly higher than that of most low-carbon options in the power sector.

Using electrolytic hydrogen powered by low-carbon electricity to decarbonize the power system is unlikely to be any more appealing, unfortunately, for reasons that have to do with the concept of round-trip efficiency (RTE), defined as the percentage of energy input to a system that can later be retrieved for productive purposes. Figure 7 shows that the RTE for using electrolytic hydrogen in power generation is only 24%. In other words, 76% of the electricity used to make the hydrogen is not recovered and in practical terms can be considered lost. Put another way, in a grid that is not already fully decarbonized
(and particularly one where full grid decarbonization still faces considerable hurdles), four units of clean electricity will be diverted from further grid decarbonization to deliver one unit of clean electricity, effectively losing three units of clean electricity that could be used to serve other direct electricity end uses.

Figure 7: System Efficiency (RTE) of Electrolytic Hydrogen Use in Power Generation

Long-duration energy storage – The role that electrolytic hydrogen might usefully play in a decarbonized power system is as a form of long-duration energy storage for grid balancing at times when renewable generation would otherwise exceed demand and would need to be curtailed. However, this role is likely to be relevant only in a fully decarbonized grid. Even then, an evidence-based analysis would be needed to examine the entirety of the power system design, evaluate alternatives for long-duration energy storage, and optimize total system cost and decarbonization pathways.

Applications That Lack Justification

Residential use – Numerous independent studies have concluded that alternatives such as heat pumps, solar thermal systems, and district heating are more economic, more efficient, less resource intensive, and have a smaller environmental impact compared to hydrogen when it comes to options for home heating. In addition, though hydrogen is routinely handled in industrial applications, its use in residential settings presents potentially serious safety hazards, both because of hydrogen’s high susceptibility to leakage (without a visible flame or odorant) and because its flammability range is six times that of natural gas. For these reasons, hydrogen is highly unlikely to be the preferred decarbonization option for residential use.

Light-duty vehicles – Hydrogen fuel cell vehicles require approximately 2.0–2.5 the amount of the energy as electric vehicles and their costs per kilometer or mile traveled are multiple times higher, which is likely a key factor behind their limited sales and the small number of auto manufacturers with active efforts to develop a hydrogen passenger vehicle. The advantages that light-duty fuel cell electric vehicles currently offer relative to battery electric vehicles (longer range and shorter refueling time) may be important to some users, but battery technology improvements will likely render these features less decisive in terms of favoring FCEVs.

Final Thoughts

Low-carbon hydrogen may be an essential tool for reducing greenhouse gas emissions from certain industries, but its physical and cost realities suggest it may not be an all-purpose tool (the ‘Swiss army knife’, as it were) for decarbonization, as it has sometimes been presented. Furthermore, hydrogen is not a primary source of energy and does not contribute to energy security given the costs and large amounts of energy required to produce hydrogen – indeed, these energy requirements (even more than the availability of electrolysis equipment or natural gas reformers) are what constrain low-carbon hydrogen supply today.

Hydrogen already is and will continue to be a critical chemical that must be produced on an industrial scale for all kinds of purposes, including to manufacture a variety of essential chemicals, as a feedstock for the fertilizers required to feed the global population, and to provide fuels for key segments of the transportation sector. The main debate now does not center on whether it will be technically feasible to greatly expand low-carbon hydrogen production and use, but rather where and to what extent switching to low-carbon hydrogen can be an energy-efficient and cost-effective strategy for reducing greenhouse gas emissions, taking into account the availability of alternative decarbonization pathways.

In short, policymakers will need to be thoughtful in deciding how to prioritize clean hydrogen and in designing smart policies around low-carbon hydrogen production and use. Doing so will be critical to charting a smart path to rapid and effective economy-wide decarbonization.