The implementation of efficient and cost effective thermal energy storage in concentrated solar power (CSP) applications is crucial to the wide spread adoption of the technology. The current push to high-temperature receivers enabling the use of advanced power cycles has identified solid particle receivers as a desired technology. A potential way of increasing the specific energy storage of solid particles while simultaneously reducing plant component size is to implement thermochemical energy storage (TCES) through the use of non-stoichiometric perovskite oxides. Materials such as strontium-doped lanthanum cobalt ferrites (LSCF) have been shown to have significant reducibility when cycling temperature and oxygen partial pressure of the environment [1]. The combined reducibility and heat of the oxidation and reduction reactions with the sensible change in temperature of the material leads to specific energy storage values approaching 700 kJ kg −1 . A potential thermochemical energy storage system configuration and modeling strategy is reported on, leading to a parametric study of critical operating parameters on the TCES subsystem performance. For the LSCF material operating between 500 and 900°C with oxygen partial pressure swings from ambient to 0.0001 bar, system efficiencies of 68.6% based on the net thermal energy delivered to the power cycle relative to the incident solar flux on the receiver and auxiliary power requirements, with specific energy storage of 686 kJ kg −1 are predicted. Alternatively, only cycling the temperature between 500 and 900°C without oxygen partial pressure swings results in TCES subsystem efficiencies up to 76.3% with specific energy storage of 533 kJ kg −1 .

This paper reports the synthesis, characterization and evaluation of different weight loadings of cobalt ferrite (CoFe 2 O 4 ) in 8 mol% yttria-stabilized zirconia (8YSZ) via the co-precipitation method. Prepared powders were calcined at 1350 °C for 36 hours and 1450 °C for 4 hours in air. These powders were then formed into a porous structure using sacrificial pore formation via oxidation of co-mixed graphite powder. These formed structures obtained were then characterized using thermogravimetric analysis (TGA), X-ray diffraction (XRD), high temperature X-ray diffraction (HT-XRD), scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). Brunauer-Emmett-Teller (BET) surface area analysis was performed on the most promising of the structures before being subjected to 50 thermal reduction-CO 2 oxidation (redox) cycles using TGA. Together, these results indicate that CoFe 2 O 4 -8YSZ can provide a lower reduction temperature, maintain syngas production performance from cycle to cycle, and enhance utilization of the reactive material within the inert support in comparison to iron oxide only structures.

A two-step thermochemical water splitting cycle using a redox system of non-volatile metal oxide is one of the promising processes for converting concentrated solar high-temperature heat into clean hydrogen in sun-belt regions. In the 1st step of the cycle or the thermal reduction step, metal oxide is thermally reduced to release oxygen molecules in an inert gas atmosphere at a higher temperature above 1400°C. In the second step or the water-decomposition step at a lower temperature, the thermally-reduced metal oxide reacts with steam to produce hydrogen. As the reactive redox metal oxide materials to be capable of working below 1400°C, nickel-doped iron oxides or Ni-ferrites supported on zirconia, and non-stoichiometric cerium oxides are the promising working materials. In the present work, a series of the nickel-ferrite redox materials of monoclinic -zirconia-supported, cubic -YSZ(yttrium-stabilized zirconia)-supported, and non-supported Ni-ferries and non-stoichiometric cerium oxide were compared on reactivity for two-step thermochemical water splitting cycle. The monoclinic -zirconia-supported Ni-ferrite produced the most quantity of hydrogen in the repeated cycles when the thermal reduction step was performed for 30 min at 1400°C and the water decomposition step for 60 min at 1000°C.

A two-dimensional, multi-physics computational model based on the finite-element method is developed for simulating the process of solar thermochemical splitting of carbon dioxide (CO 2 ) using ferrites (Fe 3 O 4 /FeO) and a counter-rotating-ring receiver/recuperator or CR5, in which carbon monoxide (CO) is produced from gaseous CO 2 . The model takes into account heat transfer, gas-phase flow and multiple-species diffusion in open channels and through pores of the porous reactant layer, and redox chemical reactions at the gas/solid interfaces. Results (temperature distribution, velocity field, and species concentration contours) computed using the model in a case study are presented to illustrate model utility. The model is then employed to examine the effects of injection rates of CO 2 and argon neutral gas, respectively, on CO production rate and the extent of the product-species crossover.

We have experimentally investigated the thermochemical decomposition of carbon dioxide using pure cerium oxide fibrous structures. Experiments were conducted on-sun with a solar furnace and include two reaction steps: the thermal reduction of CeO α to CeO β between 1500°C and 1600°C, and the re-oxidation of CeO β to produce carbon monoxide under flowing carbon dioxide at temperatures between 800°C and 1200°C. A ceria-based cycle offers some advantages over similar thermochemical cycles including the reduction of sintering and volatility issues during thermal reduction, a stable crystal structure over the range of operating temperatures, and the ability for all of the material to participate in the thermochemical reactions, i.e. there is no inert support. We present experimental results indicating that pure ceria structures perform at a level comparable to ferrite-based structures with respect to material utilization and better than the ferrites with respect to the carbon monoxide production rate during the oxidation step. We also discuss the performance potential of a solar reactor that continuously produces carbon monoxide using ceria in a two-step thermochemical cycle.

The relationship among the factors determining the thermal efficiency to convert concentrated solar thermal energy to chemical energy by the O 2 -releasing reaction of the reactive ceramics in the tow-step water splitting process has been studied for the development of rotary-type solar reactors. The α O 2 -releasing reaction which has been discovered in the present study for Ni-ferrite (NiFe 2 O 4 ) has a high chemical reaction rate ( k = 0.50 mol sec −1 /m 2 ). From calculation with the heat transfer equation in terms of heat transfer through the cavity wall with thckness, d (m), and heat absorption by the endothermic process of the O 2 -releasing reaction with rate constant, k (mol sec −1 /m 2 ), it is cralified that the heat flux of 2000 kW/m 2 can be absorbed by the α O 2 -releasing reaction using the cavity wall with the thickness, d = 0.00036 m and thermal conductivity, λ = 70 WK −1 m −1 . The heat loss in the temperature swing (ΔT = 300K) for the two step water splitting of O 2 releasing step (T H = 1773K) and H 2 generation step (T L = 1473K) is only 8.3% for the α O 2 -releasing reaction. However, the O 2 releasing reaction of NiFe 2 O 4 in the β region where the reaction rate constant is around 0.0001 mol sec −1 /m 2 , the heat flux of the concentrated solar energy to be used for conversion to chemical energy becomes very low around 300 kW/m 2 ; the heat loss by the temperature swing in the two step water splitting is 87%. It is concluded that the α O 2 -releasing reaction can be used for the rotary-type solar reactor to convert the concentrated solar thermal energy to chemical energy with a high efficiency.

Metal-oxide based thermochemical cycles, such as those including a class of iron containing materials commonly known as ferrites, involve two reaction steps: a thermal reduction at temperatures up to 1600 °C driven by a solar energy input, and a lower temperature exothermic oxidation in the presence of either carbon dioxide or water. In order to maximize performance, the reactive materials must be arranged into structures that provide an effective interface for the direct absorption of concentrated solar energy and also have relatively high surface area to support rapid chemical reactions. In this paper we discuss the attributes of reactive structures for solar thermochemical processes as well as some of the fabrication techniques currently under development at Sandia National Labs. One of these structures has been demonstrated on-sun in a two step carbon dioxide splitting cycle. The results, given in this paper, indicate that performance may be improved as the fraction of the total directly illuminated surface area is increased, reducing the need to rely on conduction or convection to distribute heat throughout the material.

The transient thermal behavior of two solar receiver-reactors for hydrogen production has been modeled using Modelica/Dymola. The simulated reactors are dedicated to carry out the same chemical reactions but represent two different development stages of the project HYDROSOL and two different orders of magnitude concerning reactor size and hydrogen production capacity. The process itself is a two step thermochemical cycle, which uses mixed iron-oxides as a redox-system. The iron-oxide is coated on a ceramic substrate, which is placed inside the receiver-reactor and serves on the one hand as an absorber for solar radiation and on the other hand as the reaction zone for the chemical reaction. The process consists of a water splitting step in which hydrogen is produced and a regeneration step during which the used redox-material is being reduced. The reactor is operated between these two reaction conditions in regular intervals with alternating temperature levels of about 800 °C for the water splitting step and 1200 °C for the regeneration step. Because of this highly dynamic process and because of fluctuating solar radiation during the day, a mathematical tool was necessary to model the transient behavior of the reactor for theoretical studies. Two models have been developed for two existing receiver-reactors. One model has been set up to simulate the behavior of a small scale test reactor, which has been built and tested at the solar furnace of DLR in Cologne. Results are very promising and show that the model is able to reflect the thermal behavior of the reactor. Another model has been developed for a 100 kW th pilot reactor which was set up at the Plataforma Solar de Almeri´a in Spain. This model is based on the first model but special geometrical features had to be adapted. With this model temperatures and hydrogen production rates could be predicted.

A two-step water-splitting thermochemical cycle using redox working material of iron-based oxide (ferrite) particles has been developed for converting solar energy into hydrogen. The two-step thermochemical cycle for producing a solar hydrogen from water requires a development of a high temperature solar-specific receiver-reactor operating at 1000–1500°C. In the present work, ferrite-loaded ceramic foams with a high porosity (7 cells per linear inch) were prepared as a water splitting device by applying ferrite/zirconia particles on a MgO-partially stabilized Zirconia (MPSZ) ceramic foam. The water splitting foam device was prepared using a new method of spin coating. A spin coating method we newly employed that has advantages of shortening preparation period and reducing of the coating process in comparison to previous preparation method reported. The water-splitting foam devices, thus prepared, were examined on hydrogen productivity and reactivity through a two-step thermochemical cycle. NiFe 2 O 4 / m -ZrO 2 /MPSZ and Fe 3 O 4 / c -YSZ/MPSZ foam devices were firstly tested for thermal reduction of ferrite using the laboratory scale receiver-reactor by a sun-simulator to simulate concentrated solar radiation. Subsequently, with another quartz reactor the light-irradiated device was reacted with steam by infrared furnace. As a result, it was possible to perform cyclic reactions over several times and to produce hydrogen through thermal-reduction at 1500°C and water-decomposition at 1100–1200°C. In further experiments, the NiFe 2 O 4 / m -ZrO 2 /MPSZ foam device was successfully demonstrated in a windowed single reactor for cyclic hydrogen production by solar-simulated Xebeam irradiation with input power of 1 kW. The NiFe 2 O 4 / m -ZrO 2 /MPSZ foam device produced hydrogen of 70–190μmol per gram of device through 6 cycles and reached ferrite conversion of 60% at a maximum.

A thermochemical two-step water splitting cycle using a redox system of iron-based oxides or ferrites is one of the promising processes for converting solar energy into clean hydrogen in sunbelt regions. Fe 3 O 4 supported on YSZ (Yttrium-Stabilized Zirconia) or Fe 3 O 4 /YSZ is a promising working material for the two-step water splitting cycle. In the water splitting cycle, an iron-containing YSZ or Fe 2+ -YSZ is formed by a high-temperature reaction between Fe 3 O 4 and YSZ support at 1400°C in an inert atmosphere. The Fe 2+ -YSZ reacts with steam and generate hydrogen at 1000°C, to form Fe 3+ -YSZ that is re-activated by a thermal reduction in a separate step at 1400°C under an inert atmosphere. In the present work, the thermal reduction was performed in a higher temperature range of 1400–1500°C while the hydrolysis reaction was carried out at 1000°C. It was confirmed by XRD analysis that the cyclic redox reactions occurred based on the same reaction mechanism when using a thermal reduction temperature between 1400 and 1500°C. The conversions of Fe 3 O 4 to Fe 2+ -YSZ were 20, 26 and 47% when the thermal reduction temperature were 1400, 1450, and 1500°C respectively, indicating that the x values in the formed Fe 2+ -YSZ or Fe x 2+ Y y Zr 1− y O 2− y /2+ x were 0.08, 0.11, and 0.19 respectively, where y = 0.15. The conversions of Fe 2+ -YSZ to Fe 3+ -YSZ in the hydrolysis reaction (at 1000°C), however, decreased from 90% to 60% when the thermal reduction temperature increased from 1400 to 1500°C. As the results, the hydrogen production reactivity of Fe 3 O 4 supported on YSZ increased from 5.6 × 10 −4 to 7.5 × 10 −4 g per gram of Fe 3 O 4 /YSZ for one cycle on the cycle average by elevated thermal reduction temperature from 1400 to 1500°C.

Hydrogen production by water-splitting thermochemical cycle based on manganese ferrite /sodium carbonate reactive system is reported. Two different preparation procedures for manganese ferrite/sodium carbonate mixture were adopted and compared in terms of materials capability to cyclical hydrogen production. According to the first procedure conventionally synthesized manganese ferrite, i. e. high temperature (1250 °C) heating in Ar of carbonate/oxide precursors, was mixed with sodium carbonate. The blend was tested inside a TPD reactor using a cyclical hydrogen production/material regeneration scheme. After few cycles the mixture resulted rapidly passivated and unable to further produce hydrogen. An innovative method that avoids the high temperature synthesis of manganese ferrite is presented. This procedure consists in a set of consecutive thermal treatments of a manganese carbonate/sodium carbonate/iron oxide mixture in different environments (inert, oxidative, reducing) at temperatures not exceeding 750 °C. Such material, whose observed chemical composition consists in manganese ferrite and sodium carbonate in stoichiometric amount, is able to evolve hydrogen during 25 consecutive water-splitting cycles, with a small decrease in cyclical production efficiency.

A two-step thermochemical cycle for solar hydrogen production using mixed iron oxides as the metal oxide redox system has been investigated. A reactor concept has been developed in which the metal oxide is fixed on multi-channelled honeycomb ceramic supports capable of adsorbing solar irradiation. In the solar furnace of DLR in Cologne coated honeycomb structures were tested in a solar receiver-reactor with respect to their water splitting capability and their long term stability. The concept of this new reactor design has proven feasible and constant hydrogen production during repeated cycles has been shown. For a further optimization of the process and in order to gain reliable performance predictions more information about the process especially concerning the kinetics of the oxidation and the reduction step are essential. To examine the kinetics of the water splitting and the regeneration step a test rig has been built up on a laboratory scale. In this test rig small coated honeycombs are heated by an electric furnace. The honeycomb is placed inside a tube reactor and can be flushed with water vapour or with an inert gas. A homogeneous temperature within the sample is reached and testing conditions are reproducible. Through analysis of the product gas the hydrogen production is monitored and a reaction rate describing the hydrogen production rate per gram ferrite can be formulated. Using this test set-up, SiC honeycombs coated with a zinc-ferrite have been tested. The influences of the water splitting temperature and the water concentration on the kinetics of the water splitting step have been investigated. A mathematical approach for the reaction rate was formulated and the activation energy was calculated from the experimental data. An activation energy of 110 kJ/mole was found.

The rotary-type solar reactor has been developed for solar hydrogen production with the two-step water splitting process using the reactive ceramic (Ni, Mn-ferrite). The rotary-type reactor has the rotating tubular cylinder covered on a reactive ceramic and dual reaction cells for O 2 -releasing and H 2 -generation reactions. The successive evolutions of O 2 and H 2 gases were observed in the O 2 releasing and H 2 generation reaction cells, respectively, with the prototype (small) reactor (diameter of cylinder ; 4cm). There is an upper limit for the rate of H 2 gas evolution in the case of the prototype reactor because of the slow rotation rate in a small irradiation area. To confirm the practical operation of the rotary-type solar reactor with the two-step water splitting process for the simultaneous production of H 2 and O 2 gases, a scaled-up rotary-type solar reactor with 400cm 2 of the irradiation area was fabricated (diameter; 50cm). The scaled-up reactor made of inner and outer short tubular cylinders (stainless steel) has a quartz glass window for the irradiation of reactive ceramic coated on the inner tubular cylinder (cylindrical rotor) and reaction cells were aligned in the sharing spaces between the inner and outer short tubular cylinders with gas sealing mechanisms. In the reactor, reactive ceramic coated on the inner tubular cylinder was heated up to 1800K by using the infrared imaging lamps (solar simulator) with thermal flux = 600kW/m 2 . The solid solution between YSZ and Ni-ferrite was used as reactive ceramic for the scaled-up reactor in order to prevent from sintering at a high temperature in the O 2 -releasing reaction, since the sintering of reactive ceramic resulted in lowering the yield of H 2 gas evolution in the H 2 -generation reaction. The amounts of H 2 and O 2 gases evolved at the rotation rate of 0.3rpm were evaluated to 64cm 3 and 30cm 3 for 10min with the scaled-up rotary-type solar reactor, respectively, which were much larger than those with the prototype reactor. The simultaneous evolutions of H 2 and O 2 gases in the two-step water splitting process were repeated by employing the scaled-up reactor with the solid solution between YSZ and Ni-ferrite.

A thermochemical two-step water splitting cycle using a redox system of iron-based oxides or ferrites is one of the promising processes for converting solar energy into clean hydrogen in sunbelt regions. An iron-containing YSZ (Yttrium-Stabilized Zirconia) or Fe-YSZ is a promising working redox material for the two-step water splitting cycle. The Fe 2+ YSZ is formed by a high-temperature reaction between YSZ, and Fe 3 O 4 supported on the YSZ at 1400°C in an inert atmosphere. The Fe 2+ -YSZ reacts with steam and generate hydrogen at 1000–1100°C, to form Fe 3+ -YSZ that is re-activated by a thermal reduction in a separate step at temperatures above 1400°C under an inert atmosphere. In the present work, a ceramic foam coated with the Fe-YSZ particles is examined as the thermochemical water splitting device for use in a solardirectly-irradiated receiver/reactor system. The Fe-YSZ particles were coated on an Mg-partially-stabilized zirconia foam disk and the foam device was tested on the two-step water splitting cycle being performed alternately at temperatures between 1100 and 1400°C. The foam device was irradiated by concentrated visible light from a sun-simulator at the peak flux density of 1000 kW/m 2 and the average flux density of 470 kW/m 2 in a N 2 gas stream, and then, was reacted with steam at 1100°C while heating by an infrared furnace. Hydrogen successfully continued to be produced in the repeated cycles.

The thermal reduction of metal oxides as part of a thermochemical two-step water splitting cycle requires the development of a high temperature solar reactor operating at 1000–1500°C. Direct solar energy absorption by metal-oxide particles provides efficient heat transfer directly to the reaction site. This paper describes experimental results of a windowed thermochemical water-splitting reactor using an internally circulating fluidized bed of the reacting metal-oxide particles under direct solar irradiation. The reactor has a transparent quartz window on the top as aperture. The concentrated solar radiation passes downward through the window and directly heats the internally circulating fluidized bed of metal-oxide particles. Therefore, this reactor needs to be combined with a solar tower or beam down optics. NiFe 2 O 4 / m -ZrO 2 (Ni-ferrite supported on zirconia) particles is loaded as the working redox material in the laboratory scale reactors, and thermally reduced by concentrated Xe-beam irradiation. In a separate step, the thermally-reduced sample is oxidized back to Ni-ferrite with steam at 1000°C. As the results, the conversion of ferrite reached about 44% of maximum value in the reactor by 1kW of incident solar power. The effects of preheating temperature and particle size of NiFe 2 O 4 / m -ZrO 2 were tested for thermal reduction of internally circulating fluidized bed in this paper.

A process of synthesis of mixed iron oxides for their application to solar hydrogen production is reported. To analyze the suitability of the selected technique, different compositions in the compound Mn x Ni 1−x Fe 2 O 4−5 were prepared by varying the Ni/Mn ratio between 0 and 3. The main objective was to identify the optimal amount of dopants for hydrogen production in such a magnetite. The powders were obtained from a solution of Ni, Mn and Fe nitrates by a polymeric method based on the Pechini process and were characterized by XRD and SEM. The characterization results indicated that the magnetite is fully developed at 1200°C by a multi-step solid state reaction between the mixed oxides produced after the resin heating (α- Fe 2 O 3 and nickel-iron spinel). The particles have a polygonal morphology and are softly agglomerate. Their grain size vary with de manganese content and is about 1 micron for 0.25Mn in the mixed ferrite composition and 10 micron in the mixed ferrite without nickel. The activation endothermic step eventually resulting in an oxygen-deficient ferrite was carried out within a thermogravimetric balance. The TGA/DTA mixed magnetite analysis carried out with nitrogen as inert carrier gas showed a weight loss that can be attributed to the partial reduction of the magnetite. The weight losses and the activation temperature increases when the Ni/Mn ratio decreases, being 0.5% at 700°C and 2.57% at 900°C for Ni/Mn ratio 3 and 0 respectively. A series of experimental tests will follow at laboratory test facility with indirect and direct illumination, in order to select the most adequate operation conditions and to quantify the maximum cycle efficiency for a solarized process.