Thermochemical processes based on sulfur compounds are among the most developed systems to produce hydrogen through water splitting. Due to their operating conditions, sulfur cycles are suited to be coupled with either nuclear or solar plants for renewable hydrogen production. A description of the most promising sulfur cycles, namely the Hybrid Sulfur, the Sulfur Iodine, the Sulfur Bromine and the Sulfur Ammonia processes, is given, including the work being performed for each cycle and discussing their maturity and performance for nuclear and solar applications.
A large-scale hydrogen economy requires the production of hydrogen, to be used either as an energy carrier producing electric or mechanical work or as a chemical compound employed in chemical plants [1]. Hydrogen production requires external power input to break the bonds of the molecules containing the gas, such as water, fossil fuels and biomass. Adopting water-splitting processes, hydrogen and oxygen are the only products and a closed water cycle can be realized, producing water, again, in a fuel cell or in a combustion engine. However, the direct thermal decomposition of water happens only at high temperatures on the order of 4500 K, making the process impractical for large scale scenarios [2]. Current alternatives to the direct water splitting are electrolysis and indirect thermochemical splitting. Electrochemical water splitting is realized at temperatures close to room temperatures and achieves efficiencies on the order of 70%, requiring electric power inputs of approximately 50 kWh/kgH2 [3][4]. Recent studies projected H2 production costs on the order of 4.96–5.78 $/kg for large scale alkaline electrolysis systems installed in South Carolina (USA) [5]. High temperature steam electrolysis processes, in principle, can achieve a higher efficiency than low temperature electrolysis and are currently under investigation as a possible alternative process to be coupled with solar and nuclear power sources [6][7]. Water can also be split through heat-driven chemical reactions with recirculation of intermediate substances in the cycle [8][9][10][11][12]. Such processes, referred to as thermochemical hydrogen production cycles, have been attracting interest since the 1970s. The compounds, recirculating inside the process, are based on many different elements, such as sulfur, iodine, bromine, iron, manganese, calcium, or chlorine, depending on the specific cycle. These compounds undergo reduction reactions, producing oxygen, and oxidation reactions, producing hydrogen. Comprehensive reviews of the main thermochemical cycles, used for both nuclear and solar applications, can be found in References [13][14]. The screening analyses [13] were carried out, identifying the performance of more than 100 cycles, based on selected characteristics and targets. The performance metric included the following characteristics: number of chemical reactions, number of product separation steps, number and abundance of chemical elements, corrosiveness of the process solutions, presence and flow of solid compounds, maximum temperature, availability of cycle chemistry demonstration data and availability of data demonstrating projected efficiency and costs [13]. Among the screened thermochemical processes, sulfur-based cycles, which see a thermal sulfuric acid decomposition section in common, were among the high-ranking processes [13]. Sulfur-based thermochemical cycles do not include any solid reactant movement and operate at relatively low temperatures (700–900 °C) compared with other competing processes. In addition, the main S-based cycles include two to three main reaction sections, limiting the complexity of the overall process and reducing the product separation sections. The main sulfur-based cycles, currently under investigation, are the Hybrid Sulfur (HyS) process, the Sulfur Iodine (SI) process and the Sulfur Bromine cycle. Two additional cycles have also been examined, namely the General Atomics S-cycle and the Sulfur Ammonia cycle. The General Atomics S-cycle was one of the first cycles proposed by General Atomics in 1970s [15]. However, after initial testing, the research and development work was interrupted mainly due to the presence of five main reactions as well as some critical thermodynamics and kinetics limitations [13][16]. The Sulfur Ammonia cycle, briefly discussed in Section 2, is being studied for solar applications but still requires additional fundamental development to propose a prototype demonstration with production of hydrogen and oxygen [17].
Among the thermochemical cycles being currently examined for nuclear and solar hydrogen production, sulfur-based processes are likely the most advanced systems with high potential for low-cost and high-efficiency large scale H2 production. They are characterized by many appealing features, when compared with other thermochemical processes [13][18]. Mainly, sulfur-based cycles operate at relatively low temperatures and do not require any solid material movement.
The Hybrid Sulfur (HyS) cycle was conceived by Los Alamos National Laboratory and Westinghouse [19][20]. Currently, it is one of the main cycles (and the main sulfur-based cycle) under investigation internationally for large-scale hydrogen production. Given the operating conditions, the process can be coupled with both nuclear source and concentrating solar source.
The process splits the water molecule through the recirculation of compounds based on sulfur, oxygen and hydrogen within two main chemical sections:
H2SO4 → SO2 + H2O + ½ O2 |
(1) |
SO2 + 2 H2O → H2SO4 + H2 |
(2) |
The thermal decomposition of H2SO4 to SO2, H2O and O2 (Reaction (1)) is an endothermic process, common to each sulfur-based thermochemical cycle. After the decomposition of the sulfuric acid, sulfur dioxide and oxygen are separated at lower temperatures. Sulfur dioxide is recirculated within the thermochemical process to drive the electrochemical low temperature step (Reaction (2)), while oxygen is extracted from the cycle. Hydrogen is produced in the exothermic process (Reaction 2), through the electrochemical oxidation of SO2. The sulfuric acid is recirculated and concentrated to drive the H2SO4 thermal decomposition section. The hydrogen is separated from the other products, purified and extracted from the plant.
The electrochemical section, which is the distinguishing section of the HyS process, realizes the SO2 oxidation at an electrolyzer anode to form H2SO4 and hydrogen ions (H+), which recombine with electrons and form the hydrogen molecule at the cathode. The electrolyzer works between room temperatures and about 140 °C, mainly depending on the membrane employed in the component.
The electrochemical component is currently designed either as a liquid-fed or as a vapor-fed electrolyzer. The liquid-fed configuration was developed at the Savannah River National Laboratory (SRNL), during the U.S. Department of Energy (DOE) Nuclear Hydrogen Initiative [21]. The feeding mixture (liquid SO2 and H2O) is oxidized at the anode of the electrolyzer to form H2SO4, which feeds the thermal decomposition section, H+ protons and electrons. Protons pass through the electrolyzer membrane to the cathode and recombine with external electrons to form H2 [22][23][24]. One of the main issues associated with the use of the liquid fed electrolysis system approach was the sulfur dioxide cross over through the component membrane. Recent experimental data, collected at the SRNL, demonstrated the ability of a novel configuration employing proton exchange membrane (PEM) to prevent sulfur accumulation and sulfur dioxide crossover [25]. The vapor fed electrolysis configuration has recently been developed at University of South Carolina and SRNL [26]. Dry vapor SO2 is fed to the anode of the electrolyzer, while liquid H2O is fed to the cathode of the component. Water diffuses across the membrane to the anode because of the water activity difference and pressure gradients. Water is also transported again to the electrolyzer cathode by electro-osmotic drag. Sulfuric acid and hydrogen protons are produced at the anode. The protons then recombine with electrons to produce hydrogen at the cathode of the component [26].
The reversible cell potential of the electrochemical SO2 oxidation reaction is 0.158 V, i.e., approximately 13% of the water electrolysis potential (1.229 V). The electrolyzer is currently designed adopting a traditional PEM fuel cell approach, allowing a compact design, reduced footprints and lower cost solutions [23][24]. Recent work demonstrated that a PEM electrolyzer, employing sulfonated polybenzimidazole (s-PBI) membranes, has high potential to achieve a practical cell potential of 0.6–0.7 V at current densities on the order of 500 mA/cm2 [27][28][29] This cell potential value is equal to approximately one third of the actual water electrolysis potential (i.e., 1.8 V) [20][30]. Platinum material has been adopted as the baseline catalytic formulation to reduce the kinetics overpotential of the SO2 oxidation reaction [31]. However, recent in-situ tests showed that Au nanoparticle catalysts seem to exhibit high current densities and greater stability than Pt nanoparticle catalysts [32].
The electric input, mainly required to operate the electrochemical oxidation of the sulfur dioxide, represents approximately 20% of the thermochemical process input [22]. The thermochemical process efficiency has been assessed to be equal to (realistic) values on the order of 30–40% (based on the hydrogen LHV), depending on the process flowsheet, the layout and on the electric power generation efficiency [30][33][34][35]. Techno-economic analyses, carried out for both nuclear driven and solar driven HyS processes, identified realistic nuclear H2 production costs on the order of 5.34–6.18 $/kg [36]. Reduced costs could be achieved only with high efficiency HyS process configurations and operating with high temperature nuclear reactors [37]. Solar-driven HyS processes were examined and developed for coupling with concentrating solar power plants. Hydrogen production costs were assessed to be on the order of 2.64–7.58 $/kg depending on the characteristics of the solar plant (e.g., thermochemical efficiency, heliostat cost, solar plant efficiency, location) [22][34][38][39]. Selected solar HyS plant configurations demonstrated potential to produce hydrogen at costs that can closely approach the DOE target of 2 $/kg. [40].
The Sulfur Iodine (SI) cycle was originally developed by General Atomics [13][16][41], mainly focusing on nuclear power applications.
The SI cycle has three main sections, based on compounds of sulfur, iodine, hydrogen and oxygen:
H2SO4 → SO2 + H2O + ½ O2 |
(1) |
2 HI → I2 + H2 |
(3) |
I2 + SO2 + 2 H2O → H2SO4 + 2 HI |
(4) |
The thermal decomposition of H2SO4 into SO2, H2O and O2 (Reaction 1) produces sulfur dioxide and water and oxygen. Sulfur dioxide and oxygen are recycled in an exothermic section (Reaction 4), referred to as the “Bunsen” section, while oxygen is separated from the other compounds and extracted from the plant as byproduct. Hydrogen is produced in Reaction (3), which is the distinguishing SI plant section, where hydrogen iodide (HI) decomposition to hydrogen and iodine takes place. An aqueous hydrogen iodide mixture, feeding the section at relatively low HI concentrations, requires HI concentration to operate an effective and low energy HI decomposition. The HI section is a critical step in the SI process, due to the homogeneous azeotrope in the system HI-H2O making the acid concentration and decomposition highly energy intensive and expensive [42]. Three methods have mainly been examined to concentrate and decompose the hydrogen iodide: the extractive distillation, the membrane-based distillation (i.e., electro-electrodialysis distillation) and the reactive distillation [43][44][45]. The extractive distillation concept was examined by General Atomics, proposing the I2 separation by extractive distillation using H3PO4 [41][46]. However, this approach essentially adds another compound (i.e., another section) to the cycle, requiring additional separation units, reactors and electric power input to recycle the phosphoric acid [47]. The electrodialysis distillation approach was proposed by Japan Atomic Energy Agency and Korea Institute of Energy Research [47], but still requires fundamental development for a prototype level demonstration [45][48]. The reactive distillation column approach, proposed by Knoche et al. [49], allows the separation of HI from the other compounds and the decomposition of the hydrogen iodide in a single column. This occurs by utilizing the pressure shift of azeotropic and quasi-azeotropic composition [50]. The reactive distillation approach shows some techno-economic hurdles to be overcome: (1) the column reboiler requires critical high temperature thermal power supply, which is currently provided through a heat pump that increases the plant investment cost and the lifetime costs due to the additional electric input and (2) fundamental research is still required to fully understand the reaction thermodynamics and kinetics [42][44].
The product of the HI section, after hydrogen separation, is recirculated in the Bunsen section, reacting with the H2SO4 section product, to produce sulfuric acid and hydrogen iodide and close the overall cycle. The Bunsen reaction is exothermic and occurs at temperatures on the order of 100–120 °C [13][51]. Additional work is required to demonstrate the Bunsen reaction, as well as the overall SI cycle, at a prototype level.
Realistic thermochemical efficiencies of 35–38% (based on H2 LHV) were assessed [44][52], also showing high sensitivity to the HI decomposition process configuration [42][47]. The nuclear hydrogen production cost was estimated to be on the order of 3.50–12.0 $/kg depending on the SI cycle and the nuclear reactor configuration [42][52].
The Sulfur Bromine cycle was originally developed at the Ispra facilities during the 1970s [43][53].
The cycle has three main sections, based on sulfur, bromine, hydrogen and oxygen compounds:
H2SO4 → SO2 + H2O + ½ O2 |
(1) |
2 HBr → Br2 + H2 |
(5) |
Br2 + SO2 + 2 H2O → H2SO4 + 2 HBr |
(6) |
Similarly to the other sulfur cycles, the high temperature decomposition of sulfuric acid (Reaction 1) produces oxygen and sulfur dioxide. The distinguishing section is the electrochemical decomposition of HBr (Reaction 5), producing hydrogen, separated from the other compounds and extracted from the process, and bromine. The hydrogen bromide is decomposed to bromine and hydrogen at temperatures in the range of 80–200 °C depending on the electrolyzer characteristics [43]. The electrolysis unit was developed and tested during the 1970s and 1980s, showing voltages on the order of 0.8–1.0 V at current densities of approximately 100–600 mA/cm2 [54]. The exothermic recombination of Br and SO2 (Reaction 6) occurs at temperatures in the range of 20–100 °C.
The cycle was tested producing hydrogen at 100 LH2/h for 150 h, with thermochemical process efficiencies on the order of 37% [43]. Currently, the cycle is not under investigation for nuclear or solar applications, mainly due to the high voltage required in the electrolysis unit.
The Sulfur Ammonia process was recently conceived by the Florida Solar Energy Center for solar applications [55]. The cycle represents an attempt to use both the thermal and the photonic components of the solar input. The cycle is comprised of five main reactions, based on sulfur, nitrogen, potassium, hydrogen and oxygen compounds, and requires a suitable solar system to collect both the thermal input and the solar light input:
SO2 + 2NH3 + H2O → (NH4)2SO3 |
(7) |
(NH4)2SO3 + H2O → (NH4)2SO4 + H2 |
(8) |
(NH4)2SO4 + K2SO4 → K2S2O7 + 2NH3 + H2O |
(9) |
K2S2O7 → K2SO4 + SO3 |
(10) |
SO3 → SO2 + ½ O2 |
(11) |
The cycle distinguishing reaction (Reaction 8) is the photocatalytic production of hydrogen and ammonium sulfate from an aqueous ammonium sulfite solution, occurring above room temperature (approximately 80–150 °C) and low pressures [56][57]. A sub-cycle is identified by the Reaction (9) and Reaction (10). Ammonium sulfate product is reacted with potassium sulfate (Reaction 9), at temperature on the order of 400 °C, to generate potassium pyrosulfate, which is decomposed (Reaction 10) at temperatures on the order of 550 °C, to K2SO4 and SO3 [57]. K2SO4 is recirculated to drive Reaction (9) and close the sub-cycle. SO3 is catalytically decomposed to SO2 and O2 (Reaction 11), as for the other sulfur-based thermochemical processes, at high temperatures on the order of 800–1000 °C. The separation of sulfur dioxide and oxygen occurs in Reaction (7) with water mixing and production of aqueous ammonium sulfite. Thermodynamic and kinetics tests demonstrated the feasibility of the proposed cycle at laboratory scale [56], but additional work is required to demonstrate the actual performance of a closed cycle at prototype level.
Every sulfur-based thermochemical cycle is comprised of a high temperature section, referred to as H2SO4 thermal section, where sulfuric acid is decomposed to sulfur dioxide, oxygen and water. The sulfur dioxide and the water are separated from the oxygen, which is extracted from the cycle as byproduct, and processed in the other sections of the thermochemical cycles. The main processes occurring in the H2SO4 thermal section are: (1) high temperature decomposition of concentrated sulfuric acid, (2) concentration of the inlet sulfuric acid mixture, (3) separation of the oxygen from the sulfur dioxide mixture. The oxygen is then purified, reaching the required purity targets, and extracted from the plant as byproduct.
The sulfuric acid decomposition section is the highest temperature section of each sulfur-based thermochemical process, where the endothermic decomposition of H2SO4 to SO2, H2O and O2 occurs. The H2SO4 decomposition takes place in two separate steps. The first reaction sees the vaporization and the instantaneous (i.e., equilibrium) decomposition of the H2SO4 mixture, feeding the section, to SO3 and H2O, as shown in Reaction (12), at temperatures in the range of 300–450 °C:
H2SO4 → SO3 + H2O |
(12) |
The second reaction sees the decomposition of the products of Equation (12) to SO2 and O2, as shown in Equation (13):
SO3 + H2O → SO2 + H2O + ½ O2 |
(13) |
The endothermic reaction of Equation (13) is thermodynamically favored at high temperatures and low pressures. For a reasonable conversion, this reaction should take place at temperatures usually on the order of 750–900 °C. The feeding sulfuric acid mixture is at temperatures of approximately 200–300 °C and concentrations on the order of 70–90 wt %, depending on the plant configuration and the operating conditions.
Ceramic materials, namely silicon carbide (SiC) or siliconized SiC (Si-SiC), were found to be the most (and likely only) suitable materials for the high temperature sulfuric acid vaporization and decomposition, given the operating temperatures and the aggressive environment [58][59].
The decomposition of SO3 to SO2 (Equation (13)) requires the use of a suitable catalyst to achieve proper SO2 yield. Pt supported catalyst, namely Pt formulations on TiO2 support, seems to represent the baseline catalytic formulation for the SO3 decomposition reaction [60]. Recent studies also demonstrated the feasibility of metal oxide catalysts, namely Fe2O3, CuO and Fe–Cr, retaining their chemical and structural stability after exposure to reaction conditions [59]. However metal oxide formulations seem to be good candidates for very high temperature reactions, while they suffered from sintering phenomena and catalytic activity reduction at temperatures lower than 800 °C [61].
Two main reactor concepts have been proposed to operate the H2SO4 decomposition. The first reactor concept, which sees a direct coupling of the H2SO4 decomposition process with the external thermal source, is referred to as a direct cavity reactor, and is especially suited for solar applications [62][63]. The second concept is referred to as an indirect solar tubular reactor, e.g. adopting a reactive bayonet heat exchanger, and is based on an indirect coupling with the external source [64][65]. In the first configuration, the thermal power is provided directly by the primary source without the presence of intermediate heat exchanger loops. The second concept includes an intermediate heat transfer fluid, transferring the required thermal input and exchanging it with the H2SO4 mixture.
This entry is adapted from the peer-reviewed paper 10.3390/pr8111383