Ammonia in Hydrogen Economy
Ammonia in Hydrogen Economy
IN A NET-ZERO
HYDROGEN
ECONOMY
March 2023
Oscar Serpell,
Zakaria Hsain,
Amy Chu,
and Walter Johnsen
1
TABLE 1: COMPARISON OF ENERGY (KILOWATT HOURS) PER UNIT VOLUME (LITERS) OF BATTERIES, HYDROGEN, AND AMMONIA
Lithium Ion Battery Liquid Ammonia Hydrogen at 1 Hydrogen at 300 Hydrogen at 700 Liquefied
bar, 300 K bar, 300 K bar, 300 K Hydrogen
0.45 MWh/m3 3.58 MWh/m3 0.003 MWh/m3 0.67 MWh/m3 1.34 MWh/m3 2.3 MWh/m3
Existing lithium-ion cells can be permanently damaged leveraging either carbon capture and sequestration
if left uncharged and will slowly lose charge over time, (CCS) or carbon-neutral electricity or heat.
limiting their usefulness for long-duration storage cycles.
Electrolysis, the splitting of water into hydrogen and
Electrochemical batteries are also heavy and have
oxygen, is one of the most promising methods of
relatively low energy density (>0.5 kWh/L), making many
producing hydrogen without the use of fossil fuels and will
applications such as in long-distance commercial air
be the assumed production method for this digest. The
travel completely impractical.
analysis could easily be replicated assuming an alternative
Hydrogen could serve as exactly the complementary method of carbon-neutral hydrogen production.
storage solution that we need for deep decarbonization.
Renewable electrolysis allows hydrogen to effectively
serve as a stable store of energy which can then be
converted back to electricity using hydrogen fuel cells or THE LIMITATIONS OF HYDROGEN
thermal energy through combustion.
Unlike batteries, which need to be transported by road or Unlike fossil fuels, which are extracted as an energy-
rail, hydrogen could, in theory, be moved long distances producing natural resource, hydrogen must be synthesized.
through pipelines or compressed as a lightweight fuel This is because it occurs naturally in low atmospheric
alternative for air travel. Because it does not degrade concentrations and capturing hydrogen is not economically
from disuse, hydrogen could be a favorable alternative to viable for the scale required in the energy sector. Instead,
lithium-ion cells in long-duration storage applications. it is produced by breaking down other molecules like
methane or water using industrial processes.
Energy yield from hydrogen combustion is similarly can be then extracted when and where it is needed by
limited by the same maximum energy output heating ammonia to high temperatures.
(approximately 33.33 kWh/kg). Whereas traditional
Despite enthusiasm for ammonia as an energy vector,
ICEs can scrape past 50% efficiency, hydrogen ICEs
there has been limited analysis of the energy requirements
achieve about half that efficiency making them largely
and situational suitability of this additional phase in
impractical as a clean transportation solution.
the hydrogen storage cycle. Because carbon-neutral
Since they are converting heat energy into kinetic production of hydrogen is already so energy intensive,
energy, these efficiencies, combined with the energy any additional energy needed for storage, extraction,
requirements of electrolysis result in a roundtrip and transportation may jeopardize the scalability and
efficiency of approximately 42% for fuel cells and just sustainability of this proposed storage solution.
17% for hydrogen powered combustion engines.
A process to store renewable hydrogen in ammonia Per this expression, 1,000 cubic meters (m3) of liquefied
could neutralize many of the storage challenges of hydrogen stored for 100 days would be reduced to
hydrogen. Hydrogen can be stored in ammonia through 594 m3 solely due to boil-off losses. Boil-off losses,
the Haber-Bosch process, which combines hydrogen combined with the energy used to liquefy the hydrogen
gas with nitrogen gas to make ammonia. Hydrogen in the first place, results in the short-term (7 days)
4 kleinmanenergy.upenn.edu
storage efficiency of 53%, and a storage efficiency of FIGURE 1: NET REMAINING ENERGY OF HYDROGEN AND AMMONIA
just 21% in seasonal storage applications (182 days), as STORAGE OVER TIME
shown in Figure 1.
process. One possibility is to use hydrogen obtained require 1.17–1.76 MWh/t-NH3 (6.63–9.97 MWh/t-H2)
through electrolysis instead of SMR. The theoretical (Salmon and Bañares-Alcántara 2021).
minimum for such a process is 5.92 MWh/t-NH3 (33.5
MWh/t-H2) (Smith et al. 2020). However, current best
available technologies have efficiencies closer to 10–12
MWh/t-NH3 (56.7–68 MWh/t-H2) including the energy COMPARING ENERGY EFFICIENCIES
required for electrolysis (Giddey et al. 2017).
Extraction of hydrogen from ammonia can be achieved Generally, compression of hydrogen, rather than
through the ammonia cracking process, 2NH3 N2 + liquefaction of ammonia or hydrogen, is the more energy
3H2. This is also an energy intensive process, typically efficient method of storing hydrogen, if space and
requiring temperatures exceeding 900ºC (Giddey et al. transportation are not limiting factors. This is attributable
2017). The high temperatures are required to give high to the boil-off rates associated with hydrogen
yields of hydrogen. liquefaction and the energy costs of ammonia synthesis
and cracking. Compressed hydrogen still requires
Lower temperatures afford larger concentrations of
energy for the initial compression process, but once
unreacted ammonia and make the utilization of the
it is fully sealed and stable within appropriately-rated
product hydrogen challenging due to the high purity
canisters or pipelines, it has effectively zero losses.
demands of fuel cells. The theoretical minimum
required for the cracking process is around 0.88
MWh/t-NH3 (5.0 MWh/t-H2 released) (Thomas and
Parks 2006), though currently available technologies
Electric Energy The energy demands in MWh per metric ton of hydrogen for each step (1) Dias et al. 2020 (7) El-Breiki and Bicer 2020
along the hydrogen storage processes using three different techniques: (2) Ni et al. 2006 (8) Giddey et al. 2017
Hydrogen (H2) compressed hydrogen, liquefied hydrogen, and liquefied ammonia. (3) Blain 2022 (9) U.S. DOE 2015
Ammonia (NH3) (4) Tolga et al. 2010 (10) Hosseini and Butler 2019
(5) Sheffield et al. 2014 (11) Rohland et al. 1992
(6) Gardiner 2009
6 kleinmanenergy.upenn.edu
Overall Return on Energy Investment 63–84% 42.5– 52.3% 41.8–51% 52.5–66% 38.5%–45%
Shows the total energy input requirements for five scenarios of hydrogen synthesis, storage, transportation, and consumption for the purposes of storing renewable electricity using compressed or
liquefied hydrogen of liquefied ammonia. The total energy required to produce and store 1 metric ton of hydrogen is given in row five with values ranging from 39.7 MWh/t-H2 in Scenario 1 to 86.5
MWh/t-H2 in Scenario 5.
Both ammonia and hydrogen liquefaction experience energy demands for ammonia synthesis and cracking
constant boil-off, reducing the roundtrip efficiency of are overtaken by the combined energy demands of
these pathways over time. However, while the boil-off electrolysis and liquefied hydrogen storage.
rate of ammonia only reduces the stored energy by
However, slight variations in the assumed efficiencies
a small fraction over a six-month storage period, the
of reactions or changes in the boil-off rate for hydrogen
overall quantity of liquefied hydrogen falls by close to
or ammonia could significantly impact the time frame for
60% over that same period.
this break-even point. The efficiency ranges in Table 2
Furthermore, ammonia can be liquefied using only a demonstrate the difficulty in precisely calculating the return
small fraction of the stored energy, while liquefying on energy invested for any of the five storage scenarios.
hydrogen requires almost 50% of the stored energy.
On the other hand, using ammonia directly as an energy
This comparative advantage of ammonia is somewhat
source through combustion produces toxic NOx gasses,
offset, however, by the synthesis and cracking of
necessitating NOx scrubbers at any point of direct
ammonia. Ultimately, the roundtrip energy efficiency
combustion. Current studies on ammonia combustion
of liquefied ammonia and hydrogen for 182 days is
have widely explored using ammonia fuel blends with
comparable, with range overlaps between 41.8–45%
hydrogen or traditional fossil fuels in combustion engines.
If a liquid fuel is required, the choice between liquefied
In general, lower purities of ammonia creates more
hydrogen and ammonia will depend on a number of factors
CO2 and NOx emissions, and distributed scrubbing
including the efficiency of electrolysis, ammonia synthesis,
would be difficult to regulate (Erdemir 2020). It is
and ammonia cracking, the expected duration of storage,
worth mentioning that while energy generation through
and the available storage and transportation infrastructure.
combustion of 100% ammonia eliminates the possibility
If we assume a 40 MWh/t-H2 energy cost of electrolysis, of CO2 emissions, it may still produce NOx due to the
a 60 MWh/t-H2 cost of renewable ammonia synthesis, partial oxidation of ammonia or the product nitrogen gas.
and 8 kWh/t- H2 for ammonia cracking, we can use the
boil-off rates of both fuels to calculate a break-even
point of 11 days. Before this point, liquefied hydrogen is
more energy efficient than ammonia. After this point, the
Ammonia’s Role in a Net-zero Hydrogen Economy 7
TAPPING INTO PIPELINE INFRASTRUCTURE The repurposing of existing natural gas pipelines to
transport carbon-neutral alternative fuels offers an
invaluable opportunity to bypass many of these new
infrastructure challenges. The energy transition will require
Independent of the roundtrip efficiency of each fuel, one
enormous infrastructure investments, but any opportunity to
of the most commonly stated advantages of ammonia
utilize existing infrastructure can save time and cut costs.
over compressed hydrogen or liquefied hydrogen is that it
could benefit from the extensive existing ammonia pipeline Repurposing natural gas pipelines to transport hydrogen
infrastructure used primarily by the agriculture sector. would require extensive modifications, which include
replacing components such as compressors, valves,
The United States, for example, has a network of more than
pressure regulators, and sealing membranes to conform
5,000 kilometers of pipeline in operation today. Recently,
with the flow properties of hydrogen. Aging pipelines
this pipeline network shrank with the decommissioning
that are at or near the end of their lifetimes would need
of the 1,800-km Magellan pipeline, which connected the
to be replaced.
Southern Plains and Western Corn Belt with production
facilities in Texas and Oklahoma; however, it still exceeds Additionally, as hydrogen is known to cause embrittlement
existing hydrogen transport infrastructure (~2,500 km in and premature cracking in steel, it might be more prudent
the United States) (Bouwkamp et al. 2017). to use novel embrittlement-resistant grades of steel or
fiber-reinforced polymers in any new pipeline construction
For comparison’s sake, the U.S. has 490,000 kilometers
(Bouwkamp et al. 2017). Existing pipelines that are deemed
of high-pressure natural gas transmission pipelines and
safe to carry hydrogen could benefit from a polymer
over 3 million kilometers of lower-pressure natural gas
coating on their inner walls to enhance their resistance to
distribution pipelines for last-mile delivery to end users
embrittlement and, thus, their operating lifetime.
(Hydrogen Council and McKinsey 2021). Neither the
existing hydrogen nor the existing ammonia infrastructure In financial terms, retrofitting existing land transmission
comes anywhere close to meeting the demands of a pipelines to carry hydrogen could cost between $0.6 and
heavily hydrogen-dependent energy system, and therefore $1.2 million per kilometer, while the cost of constructing
does not offer either fuel a significant leg up. new pipelines could be between $2.2 and $4.5 million
per kilometer (Hydrogen Council and McKinsey 2021).
Globally, demand for ammonia is equivalent to 176 million
metric tons per year. If hydrogen were to replace the energy Retrofitting existing natural gas infrastructure for
demands of just 50% of present-day natural gas, it would ammonia may be significantly more involved, since the
require at least 3.49 billion metric tons of ammonia per characteristics of existing ammonia pipelines are quite
year. This represents a 20-fold increase in global ammonia distinct from natural gas pipelines. Whereas natural gas
production and a similar increase in available transportation transmission pipelines are typically pressurized to 500–
and storage infrastructure. 1,200 psi, ammonia pipelines typically operate at just 250
psi. At this pressure, ammonia is a relatively heavy liquid.
The story is the same for hydrogen. Today, global hydrogen
production is approximately 74 million metric tons. To meet This means that if natural gas pipelines were to be
50% of the energy demand currently met by natural gas repurposed for ammonia transport, they would either
would require at least 617 million metric tons per year—an need to be adapted to operate at much lower pressures
increase of more than 800%. or under a much higher weight burden. On the other
hand, liquid ammonia is non-corrosive and does not
Arguably, the true infrastructure advantage lies in how
exhibit the same embrittlement properties as hydrogen,
these two energy vectors can make use of existing natural
meaning that materials restrictions are not as stringent.
gas infrastructure. Fuel pipelines are notoriously difficult
to complete. They are expensive, require collaboration
across multiple jurisdictions, and often meet fierce public
resistance and legal roadblocks.
8 kleinmanenergy.upenn.edu
Black & Veatch. 2020. “Hybrid LNG & Ammonia Infrastructure: Key to a Green Economy.” Zakaria Hsain is a postdoctoral researcher in the
https://doi.org/10.31826/9781463236762-toc.
Pikul Group at the University of Pennsylvania. He
Bouwkamp, Nico, Al Burgunder, Dan Casey, Amgad Elgowainy, Leah Fisher, Jim Merritt,
Eric Miller, et al. 2017. “Hydrogen Delivery Technical Team Roadmap.” https://www.energy. holds a Ph.D. in mechanical engineering and applied
gov/sites/prod/files/2017/08/f36/hdtt_roadmap _July2017.pdf.
mechanics from Penn and is a former Kleinman Center
Dias, Véronique, Maxime Pochet, Francesco Contino, and Hervé Jeanmart. 2020. “Energy
and Economic Costs of Chemical Storage.” Frontiers in Mechanical Engineering 6. https:// research assistant.
doi.org/10.3389/fmech.2020.00021.
EPRI, NYPA, and GE. 2022. “Hydrogen Cofiring Demonstration At New York Power Wan-yi “Amy” Chu is an Assistant Professor of
Authority’s Brentwood Site: GE LM6000 Gas Turbine.” https://restservice.epri.com/
publicdownload/000000003002025166/0/Product. chemistry at St. Mary’s College of California and a
Erdemir, Dogan, Ibrahim Dincer. 2020. “A Perspective on the Use of Ammonia as a Clean former postdoctoral researcher in the Goldberg
Fuel: Challenges and Solutions.” International Journal of Energy Research 45 (4). https://
doi.org/10.1002/er.6232. Group located in the Department of Chemistry at
Folga, S.M. 2007. “Natural Gas Pipeline Technology Overview.” the University of Pennsylvania.
Giddey, S., Badwal, S. P. S., Munnings, C., and Dolan, M. 2017. “Ammonia as a Renewable
Energy Transportation Media.” ACS Sustainable Chemistry & Engineering 5: 10231–10239 Walter Johnsen is a Ph.D. candidate in the department
Hernandez, Drake D., and Emre Gençer. 2021. “Laying the Regulatory Groundwork for
Hydrogen in the United States.” Utility Dive. https://www.utilitydive.com/news/laying-the-
of chemistry and the 2022 Kleinman Philadelphia Energy
regulatory-groundwork-for-hydrogen-in-the-united-states/601408/. Authority Fellow. He is also a member of the Kleinman
Hydrogen Council, and McKinsey & Company. 2021. “Hydrogen Insights.” https://
hydrogencouncil.com/wp-content/uploads/2021/02/Hydrogen-Insights-2021.pdf.
Center Student Advisory Council.
Langeraar, Jan Willem, Marcel Keezer, and Ruud Bakker. 2019. “Workshop: Challenges
for the Production and Application of Liquefied Hydrogen.” https://prod5.assets-cdn.io/ Cover Photo: istock.com/kodda
event/3757/assets/8431103583-8e79cc08e1.pdf.
Loginow, A.W., and E.H. Phelps. 1962. “Stress-Corrosion Cracking of Steels in Agricultural
Ammonia.” Corrosion 18.
Mauler, Lukas, Fabian Duffner, Wolfgang G. Zeier, and Jens Leker. 2021. “Battery Cost
Forecasting: A Review of Methods and Results with an Outlook to 2050.” Energy &
Environmental Science 14: 4712–4729
Rohland, B., J. Nitsch, and H. Wendt. 1992. “Hydrogen and Fuel Cells: The Clean Energy
System.” Journal of Power Sources 37 (1–2): 271–77. https://doi.org/10.1016/0378-
7753(92)80084-O.
The Royal Society. 2020. “Ammonia: Zero-Carbon Fertilizer, Fuel and Energy Store.”
https://royalsociety.org/-/media/policy/projects/green-ammonia/green-ammonia-policy-
briefing.pdf.
Smith, Collin, Alfred K. Hill, and Laura Torrente-Murciano. 2020. “Current and Future Role
of Haber-Bosch Ammonia in a Carbon-Free Energy Landscape.” Energy & Environmental
Science 13: 331–344
Topolski, Kevin, Evan P Reznicek, Burcin Cakir Erdener, Chris W San Marchi, Joseph A
Ronevich, Lisa Fring, Kevin Simmons, Omar Jose Guerra Fernandez, Bri-Mathias Hodge,
and Mark Chung. 2022. “Hydrogen Blending into Natural Gas Pipeline Infrastructure:
Review of the State of Technology.” Golden, CO. https://www.nrel.gov/docs/
fy23osti/81704.pdf.
University of Pennsylvania
Stuart Weitzman School of Design
Fisher Fine Arts Building, Suite 401
220 S. 34th St.
Philadelphia, PA 19104
P 215.898.8502
F 215.573.1650