Argonne National Laboratory (ANL) is developing low temperature thermochemical cycles designed to split water and produce hydrogen at 550°C or less. The rationale for this R&D effort is to identify new technologies that can produce hydrogen cost effectively without greenhouse gas emissions. Over three hundred thermochemical cycles have been reported in the literature. Most of these cycles use aggressive chemicals and require high temperatures, 850°C and above. Engineering and materials issues are therefore extremely challenging. Lower temperature cycles may reduce the thermal burden and mitigate demands on materials as well as on potential heat sources. ANL has identified the copper-chlorine (Cu-Cl) cycle as one of the most promising cycles for the lower temperature region on the basis of successful proof-of-principle work. A simplified representation of the cycle is shown in Table 1 along with the reaction temperatures used in the proof-of-principle work. The kinetics for the three reactions were reasonable. The maximum temperature of the cycle was 530°C and experimental yields for the three thermal reactions (#1, 3, and 4 in Table 1) were high. Competing reactions were not observed in the hydrogen and oxygen generation reactions (#1 and 4, respectively). However, examination of the products of the hydrolysis reaction of CuCl₂ (#3) showed CuCl and CuO·CuCl₂ in the solid phase and both HCl and Cl₂ in the gas phase. These additional products indicate the presence of a competing reaction, the decomposition of CuCl₂. The electrochemical reaction has not been extensively studied but copper deposition was observed with cell potentials between 0.4 and 0.6 V.

#	Reaction Stoichiometry	Temperature (°C)
1	2Cu + 2HCl(g)® 2 CuCl(l) + H2(g)	425-450
2	4CuCl(s) ® 2CuCl2(a) + 2Cu	<100
3	2CuCl2(s) + H2O(g) ® CuO·CuCl2 (s) + 2HCl(g)	300-375
4	CuO·CuCl2 (s) ® 2CuCl(l) + ½O2(g)	450-530

In addition to the experimental work, a simulation was prepared for the three thermal reactions using Aspen Plus. The 2005 model used reaction blocks modeled with stoichiometric reactors. A simulation with parallel equilibrium reactors confirmed that the yields from the stoichiometric reactors substantially matched the equilibrium concentrations for hydrogen and oxygen generation. Process conditions, such as pressures, temperatures, and feed concentrations, were systematically varied in order to find conditions where the yields were close to 100%, as desired. Since only estimated thermodynamic data were available for CuO·CuCl₂, the simulation forced the formation of this compound for a yield of 100%. Other assumptions in this model were the following:

1. H₂ and O₂ generation reactions go to 100% completion by choice of reaction conditions; H₂ is discharged at the separator at 23 bar (338 psi)
2. The molar ratio of water to CuCl₂ is 10 to 1 in the hydrolysis reaction run at 510°C;
3. The electrochemical cell's potential is 0.4V;
4. Efficiency of converting heat to electricity is 50%;
5. The driving force in the pinch analysis for optimizing energy usage is 10°C;
6. The maximum temperature was 550°C.

The efficiency calculated from this model was 44% (LHV), which justified continued R&D effort. Sensitivity studies using the Aspen model showed that the yield for HCl (and indirectly that for CuO·CuCl₂) was especially sensitive to the value assumed for the free energy of formation of CuO·CuCl₂ as well as the reaction temperature. For example, an uncertainty of ±11 kcal/mol was sufficient to change the yield of HCl by a factor of 7 at 300°C.

The results of the simulation and the experimental work indicated that of the three thermal reactions, the hydrolysis reaction was the least understood. Further investigation of this reaction is probably the most important for determining whether this cycle can produce hydrogen cost effectively. Low yields of CuO·CuCl₂ and corresponding high yields of CuCl and Cl₂ would significantly increase recycle, which is expensive, and lead to a difficult separation of HCl and Cl₂. Our new work was therefore concerned with optimizing the hydrolysis reaction using both modeling and experimental work.

In order to prepare a more robust simulation and to better optimize the hydrolysis reaction, we needed measured values of the enthalpy of formation, the free energy of formation and the heat capacity as a function of temperature for CuO·CuCl₂. Since CuO·CuCl₂ is not commercially available, a new method for synthesizing it was developed. This material was characterized by analysis of its chloride content (titration with AgNO₃), by x-ray diffraction, and by Raman spectroscopy. These characterizations indicated a purity of 99+%. The enthalpy of formation at 25°C was measured using two different methods and compared with the literature data. A value of 380 ± 3 kJ/mol was determined to be the most reliable. The heat capacity was measured over three temperature regions: (1) from about 4 (liquid He temperature) to 64 K (liquid N₂ temperature), (2) from 64 to 360K, and (3) from 298 to 700 K. The low temperature heat capacities were used to calculate the entropy. The free energy of formation was then derived from the experimental values for the enthalpy of formation and entropy values. All other thermodynamic data for the various chemical species were checked against literature values and the most reliable selected for inclusion in the physical property database. These data were checked for consistency. The new thermodynamic data are being added to the physical properties database. Sensitivity studies will be reexamined and the model updated. These results will be described.

New experimental work is concerned with the hydrolysis of CuCl₂ to form HCl and the CuO·CuCl₂. A new reactor system was fabricated. Experimental variables, e.g., temperature, flow rate, humidity, particle size of the CuCl₂, reactor design, were varied to maximize the production of HCl and to minimize the competing CuCl₂ decomposition reaction. This work is ongoing and its status will be discussed.

This cycle is well matched to several of the reactor concepts described in the Global Nuclear Energy Partnership, such as the Na-cooled reactor, the supercritical water reactor, and the gas reactor. Since the heat source is fossil-fuel free, the overall process does not lead to greenhouse gas emissions.