Overview of Thermochemical Water Splitting Cycles for Nuclear Hydrogen Production

Introduction

Fig. 1: Reaction scheme and schematic block diagram of the sulfur-iodine water-splitting cycle. [10,15] (Image Source: A. Fontani Herreros).

As the global demand for clean energy solutions grows, and the need to reduce greenhouse gas emissions intensifies, hydrogen has emerged as a promising alternative energy-carrier to fossil fuels. Hydrogen is already an important commodity chemical used for the production of ammonia and other petrochemicals, and it can be used to replace fossil fuels as a clean fuel in transportation and industry without emitting CO₂ during combustion. [1] However, currently, over 85% of hydrogen is produced via steam methane reforming (SMR) of natural gas, a process which uses fossil fuels directly emits carbon dioxide (CH₄ + 2 H₂O → 4 H₂ + CO₂). [1] Thus, increasing the role of hydrogen as a clean energy carrier will require substantial expansion in clean hydrogen production processes that minimize associated greenhouse gas emissions. The generation of hydrogen from nuclear energy has emerged as a potential avenue for sustainable hydrogen production, providing a carbon-free source of energy for the process, which can also be used as a means of storing the relatively constant energy output of a nuclear reactor during low-demand hours. [2]

Currently, water-splitting via electrolysis is the primary alternative to SMR for carbon-free hydrogen production, but is very energy and capital-intensive and often requires the use of expensive precious metal catalysts. [3,4] Furthermore, using nuclear electricity to electrolyze water for hydrogen production has fundamental thermodynamic limitations, since the best water electrolyzers currently have electricity-to-hydrogen conversion efficiencies of around 70%, which, when coupled to a reactor with 30-40% electricity generation efficiency, results in a maximum overall hydrogen conversion efficiency of 25-35%. [4] As a result of these limitations, researchers have explored thermochemical water splitting as an alternative means of hydrogen production. Since it is capable of using the heat output of a nuclear reactor directly, it can attain theoretical heat-to-hydrogens efficiencies of 50% or more. [4]

Thermochemical water splitting entails the use of heat to split water into hydrogen and oxygen (2 H₂O → 2 H₂ + O₂). Thermochemical water splitting cycles were first studied in the 1960s and 70s in the wake of the OPEC oil embargo, but have recently gained renewed interest as a means of sustainable hydrogen production using heat from nuclear, solar, and other low-carbon sources. [5] If heated to high enough temperatures, water will directly thermolyze into a mixture of hydrogen and oxygen atoms (including some H₂ and O₂). However, this requires extremely high temperatures (in excess of 2500°C), incurring a significant energy requirement. Thus, thermochemical water splitting instead utilizes a cycle of thermally-driven chemical reactions to indirectly separate water into hydrogen and oxygen while requiring much lower temperatures. [4] Energy is input as heat into endothermic chemical reactions, which react with water to produce intermediates that eventually dissociate into hydrogen and oxygen. Over the years, various water-splitting cycles have been proposed, utilizing a wide variety of chemical reactions. This report will provide an overview and comparison of the major types of thermochemical water splitting cycles, their various advantages and disadvantages, and the overall outlook of nuclear thermochemical hydrogen production.

Sulfur-Iodine Cycle

One of the earliest thermochemical water splitting cycles proposed was the sulfur-iodine (S-I) cycle, invented by the American corporation General Atomics in the mid-1970s. [4] In this process, iodine and sulfur dioxide are added to water, which forms hydrogen iodide and sulfuric acid, which are then separated and decomposed at high temperatures (400°C-850°C) to release oxygen and hydrogen, while generating sulfur dioxide and iodine that is re-used. [4] The reaction scheme is shown below (and also in Fig. 1):

Fig. 2: Reaction scheme and schematic flow diagram of the cerium oxide process, an example of a 2-step metal oxide water-splitting cycle. (Courtesy of the DOE.)

Overall, this reaction benefits from the use of low-cost reagents, and boasts reported hydrogen conversion efficiencies above 50% (based on heat input to LHV-based H₂ output). [6] The S-I also has the advantage of technological experience, with extensive research and testing have been conducted on the process since the 1970s, including some pilot-scale demonstrations by both General Atomics and the Japan Atomic Energy Agency (JAEA), leading to many hundreds of hours of operational experience. [7] However, despite high efficiencies, this cycle has several drawbacks, notably requiring a minimum temperature of 700°C to operate at all, and temperatures of 900°C-1000°C in order to operate at appreciable efficiencies, well above the output temperature of current light-water reactors. [8] Thus, implementation of the sulfur-iodine cycle will require either concentrated heat sources such as solar, or if used with nuclear energy, operation with very high-temperature reactors (VHTRs) or other generation IV designs that can provide the high grade of heat required for this process. [7] Additional challenges also include managing the use of highly corrosive and acidic reagents (sulfuric and hydroiodic acid) at high temperatures, the added complexity of a 3-step reaction cycle, and limited separation and recovery of intermediates, particularly hydroiodic acid (HI). [2]

Metal-Oxide-based Cycles

A second class of thermochemical water-splitting cycles utilize thermally-driven oxidation and reduction of metals and their oxides. [9] The general reaction scheme occurs in a two-step process and is described by the following reactions [3]:

The reduced species can either be a pure metal or a low-valence oxide, while the oxidized species is typically a high valence oxide. Over the years, a variety of two-step water-splitting cycles based on metal-oxides have been proposed, including cycles based on ZnO/Zn, Fe₃O₄/Fe, SnO₂/SnO, CeO₂/Ce₂O₃, Mn₂O₃/MnO, Co₃O₄/CoO, CdO/Cd, and GeO₂/GeO. [8,9] Particular characteristics vary, but overall, the metal-oxide water splitting cycles benefit from low reagent cost, fairly chemically stable intermediates with facile recovery/reuse, and a simple 2-step reaction scheme (Fig. 2). Estimates on efficiencies vary, but ZnO/Zn systems exhibit the highest theoretical efficiencies, with most demonstrated systems exhibiting efficiencies of around 50%. [10] However, they face the drawback of often requiring very high operating temperatures (1000°C-2000°C) to attain appreciable efficiencies, although the use of more specialized oxides (such as indium, or vanadium) as well as the addition of additional intermediate reaction steps can lower temperature requirements to under 1000°C. [3,11]

Calcium-Bromine Cycle

Also called UT-3, the Calcium-Bromine cycle was discovered at the University of Tokyo in 1978 by Kameyama and Yoshida and utilizes the hydrolysis and bromination of calcium and iron compounds in 4 reactions to split water into hydrogen and oxygen. [5,12] The reaction scheme occurs as follows:

Fig. 3: Reaction scheme and schematic flow diagram of the copper-chloride hybrid water-splitting cycle. (Courtesy of the DOE.)

This cycle has the advantage of requiring lower maximum temperatures (around 760°C) for hydrogen production at efficiencies comparable to the S-I cycle (30-40%). [7] However, despite a bench scale model (MASCOT) having been constructed and successfully operated for 200 h, it remains less technologically mature than the S-I cycle, and further research and development is needed to overcome issues such as HBr corrosion and reaction kinetics. [5,7]

Copper-Chlorine and Other Hybrid Cycles

The Cu-Cl cycle utilizes decomposition of metal chloride salts of various oxidation states to evolve chlorine gas and split water. Unlike the previous cycles, the copper chloride cycle involves coupling various thermal water splitting steps with an electrolysis-driven step in order to reduce the temperature requirements for the water splitting reaction. [12] The cycle can be performed in a 3-5 step process, but the 4-step process is the most commonly used and occurs in the following steps: first, copper (II) chloride is reacted with water to form a copper-oxide-chloride complex and hydrochloric acid at 400°C. This copper-oxide-chloride complex is then heated to 500°C, causing it to decompose into copper (I) chloride and oxygen gas. The copper (I) chloride solution is then electrolyzed to produce hydrogen gas at the cathode, while copper (I) chloride is oxidized back to copper (II) chloride at the anode, regenerating the starting material and closing the reaction cycle. [12,13] The overall reaction scheme is shown below and in Fig. 3.

Since electricity is used to partially drive the water-splitting cycle, the copper-chlorine cycle benefits from much lower operational temperature requirements compared to other cycles (around 500°-550°C), while still reaching experimental heat-to-H₂ efficiencies as high as 49%. [10] This temperature is compatible with existing nuclear power plant designs such as the Sodium-Cooled Fast Reactor (SFR) or the Super-Critical Water Reactor (SCWR), making it much more feasible to drive these types of cycles with nuclear energy. [2] The copper-chlorine cycle is one of several so-called hybrid water-splitting cycles that couple electrolysis with heat-driven reactions. Other notable examples include:

• Analogs of the Cu-Cl cycle that utilize other metals such as magnesium, iron, and vanadium. [8]

• The Westinghouse Hybrid sulfur cycle (HyS), a hybrid version of the S-I cycle which couples the thermally driven sulfuric acid decomposition step in the normal S-I cycle (2 H₂SO₄ + heat →  2 SO₂ + 2 H₂O + O₂) with a single electrolysis step to regenerate the sulfuric acid (SO₂ + H₂O + electricity → H₂SO₄ + H₂). [2,12] This approach has the advantage of simplifying the S-I cycle down to 2-steps rather than 3, with the drawback of requiring a specialized polymer membrane electrolyzer for the electrolysis step. [12]

• Hybrid versions of the Ca-Br cycle, which utilize the same thermal first step (CaBr₂ + H₂O → CaO + 2 HBr) and last step (CaO + Br₂ → CaBr₂ + 1/2 O₂) as the normal cycle, but use a direct electrolysis step (2 HBr + electricity → Br₂ + H₂) to regenerate bromine, allowing them to skip the intermediate iron reactions required by the non-hybrid Ca-Br cycle. [14]

In summary, hybrid cycles represent a promising approach to thermochemical water splitting due to their lower-temperature operation, but still require further development, facing challenges with corrosion, product separation, and scalability.

Conclusion and Outlook

Overall, thermochemical water-splitting represents a promising alternative to electrolysis and SMR for hydrogen production, possessing a number of potential cost and efficiency advantages. [1] Due to their high heat requirements, most water-splitting cycles are primarily being investigated for concentrated solar applications, however, the lower temperature requirements of some cycles, particularly the sulfur-iodine and hybrid copper-chlorine cycles do show potential for direct use in some high temperature fast-spectrum nuclear reactors as well as future Generation IV designs. [4] Notable agencies with ongoing research into thermochemical water-splitting cycles for nuclear hydrogen production include the Japan Atomic Energy Agency (JAEA) working on S-I and Ca-Br cycles, as well the DOE, Argonne national lab, and Atomic Energy of Canada Limited (AECL), which are working on S-I and Cu-Cl hybrid cycles for use in high-temperature supercritical water reactors. [2,8] Thus, while promising, thermochemical water splitting remains a relatively nascent technology, and no utility-scale thermochemical hydrogen production has yet been deployed at any commercial power plant worldwide. Ongoing research and pilot-scale testing will be crucial in realizing the full potential of nuclear-powered thermochemical water splitting for large-scale, sustainable hydrogen production.

© A. Fontani Herreros. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.

References

[1] N. Sazali, "Emerging Technologies By Hydrogen: A Review," Int. J. Hydrog. Energy 45, 18753 (2020).

[2] B. Yildiz et al., "Configuration and Technology Implications of Potential Nuclear Hydrogen System Applications," Argonne National Laboratory, ANL-05/30, July 2005.

[3] S. Abanades et al., "Screening of Water-Splitting Thermochemical Cycles Potentially Attractive For Hydrogen Production By Concentrated Solar Energy," Energy 31, 2805 (2006).

[4] K. R. Schultz et al., "Large-Scale Production of Hydrogen By Nuclear Energy For the Hydrogen Economy," General Atomics, GA-A24265, February 2003.

[5] S. Yalçin, "A Review of Nuclear Hydrogen Production," Int. J. Hydrog. Energy 14, 551 (1989).

[6] B. Kallman, "The Very High Temperature Reactor," Physics 241, Stanford University, Winter 2013.

[7] L. Brown et al., "High Efficiency Generation of Hydrogen Fuels Using Nuclear Power," General Atomcs, GA-A24285, June 2003.

[8] F. Safari and I. Dincer, "A Review and Comparative Evaluation of Thermochemical Water Splitting Cycles For Hydrogen Production," Energy Convers. Manag. 205, 112182 (2020).

[9] W. Mangram, "The Future of Hydrogen," Physics 241, Stanford University, Winter 2017.

[10] R. Perret, "Solar Thermochemical Hydrogen Production Research (STCH)," Sandia National Laboratory, SAND2011-3622, May 2011.

[11] B. Xu, Y. Bhawe, and M. E. Davis, "Low-Temperature, Manganese Oxide-Based, Thermochemical Water Splitting Cycle," Proc. Natl. Acad. Sci. (USA) 109, 9260 (2012).

[12] R. Pinsky et al., "Comparative Review of Hydrogen Production Technologies For Nuclear Hybrid Energy Systems," Prog. Nucl. Energy 123, 103317 (2020).

[13] M. A. Rosen, "Advances in Hydrogen Production by Thermochemical Water Decomposition: A Review," Energy 35, 1068 (2010).

[14] J. Yang, C. B. Panchal, and R. D. Doctor, "CaBr₂ Hydrolysis For HBr Production Using a Direct Sparging Contactor," Int. J. Hydrog. Energy 34, 7585 (2009).

[15] J. Riccardi et al., "Modelling of Westinghouse and Sulphur-Iodine Water Splitting Cycles for Hydrogen Production," Int. J. Energy Environ. Eng. 2, 49 (2011).