Elsevier

Energy

Volume 37, Issue 1, January 2012, Pages 591-600
Energy

Efficiency of two-step solar thermochemical non-stoichiometric redox cycles with heat recovery

https://doi.org/10.1016/j.energy.2011.10.045Get rights and content

Abstract

Improvements in the effectiveness of solid phase heat recovery and in the thermodynamic properties of metal oxides are the most important paths to achieving unprecedented thermal efficiencies of 10% and higher in non-stoichiometric solar redox reactors. In this paper, the impact of solid and gas phase heat recovery on the efficiency of a non-stoichiometric cerium dioxide-based H2O/CO2 splitting cycle realized in a solar-driven reactor are evaluated in a parametric thermodynamic analysis. Application of solid phase heat recovery to the cycling metal oxide allows for lower reduction zone operating temperatures, simplifying reactor design. An optimum temperature for metal oxide reduction results from two competing phenomena as the reduction temperature is increased: increasing re-radiation losses from the reactor aperture and decreasing heat loss due to imperfect solid phase heat recovery. Additionally, solid phase heat recovery increases the efficiency gains made possible by gas phase heat recovery.

Highlights

► Both solid and gas phase heat recovery are essential to achieve high thermal efficiency in non-stoichiometric ceria-based solar redox reactors. ► Solid phase heat recovery allows for lower reduction temperatures and increases the gains made possible by gas phase heat recovery. ► The optimum reduction temperature increases with increasing concentration ratio and decreasing solid phase heat recovery effectiveness. ► Even moderate levels of heat recovery dramatically improve reactor efficiency from 3.5% to 16%.

Introduction

Two of the most pressing challenges faced by humankind are the need to reduce anthropogenic emissions of greenhouse gases due to combustion of fossil fuels and the need to meet an expanding global energy demand. Solar energy offers an intelligent solution to both challenges. However, a transformation from fossil to solar energy requires efficient and cost-effective processes to collect, store, and transport the earth’s most abundant but diffuse and intermittent source of energy. One promising approach to harvest and store solar energy is the production of synthetic fuels via high temperature thermochemical processes. In the path discussed here, the only energy source is concentrated solar radiation and the only feedstocks are water, ideally waste or brackish water, and carbon dioxide, the effluent of combustion-based power plants or ideally CO2 recycled from the atmosphere. The solar energy is stored in chemical form as synthesis gas, a mixture of hydrogen and carbon monoxide. Synthesis gas can be used directly for electricity production and industrial applications, converted to liquid fuels like methanol or dimethyl ether using catalytic processes, or further upgraded to Fischer–Tropsch liquids or gasoline. In this manner, renewable transportation fuels—compatible with the existing infrastructure—are produced.

In the simplest chemical process, heat can be used to dissociate H2O and CO2, but the temperatures required to achieve a reasonable degree of dissociation and the lack of effective techniques to avoid recombination make direct thermal dissociation impractical [1], [2], [3], [4]. For example, 20% dissociation of CO2 and H2O at a pressure of 1 bar can only be obtained at temperatures above 2600 K and 3100 K, respectively [5]. The gaseous products tend to recombine violently. Recombination is partially avoidable with high temperature gas phase separation. However, complete separation of product gases remains a technical challenge and prohibits meaningful fuel production rates and efficiencies. To circumvent these problems, two-step solar thermochemical redox cycles have been studied for splitting H2O and CO2 to produce H2 and CO [6], [7].Solar,endothermicstep:1ΔδMxOyδox1ΔδMxOyδred+0.5O2Non-solar,exothermicstep:1ΔδMxOyδred+H2O1ΔδMxOyδox+H21ΔδMxOyδred+CO21ΔδMxOyδox+CONet:H2OH2+0.5O2CO2CO+0.5O2

The thermochemical cycle (1), (2a), (2b) replaces the single thermal dissociation reaction by two more favorable reactions at different temperatures. The net result is the splitting of water (3a) or carbon dioxide (3b). The solar step (1) is the endothermic thermal reduction of the metal oxide. This step is carried out in a receiver/reactor placed at the focal point of a solar concentrating system. The non-solar, exothermic step (2a), (2b) is the oxidation of the reduced metal oxide to produce H2 or CO. In reactions (1) and (2a), (2b), δox and δred are the non-stoichiometric coefficients of the reduced and oxidized forms of the metal oxide, and Δδ = δredδox. For integer values of δred < y and for δox = 0, reactions (1), (2a), (2b) involve a lower valence metal oxide. In the limiting case with δred = y and δox = 0, the metal oxide undergoes cycling through complete reduction and oxidation reactions. Numerous metal oxides, including Al2O3, CaO, Fe2O3, MgO, SiO2, TiO2, and ZnO, have been proposed to realize both types of cycles [6], [7], [8], [9], [10], [11]. To-date, the solar-to-fuel efficiencies of prototype reactors are low, on the order of 1%. For cycles with volatile products such as the ZnO/Zn cycle, the efficiency is limited by the irreversibilities due to the requirements for gas phase separation, which is usually accomplished by quenching the volatile decomposition products in highly-diluted streams of inert gas [12], [13].

Non-stoichiometric metal oxides realize the cycle (1), (2a), (2b) with δred and δox less than one, but have the advantage that the metal oxide remains in the solid phase throughout the cycle. Thus, non-stoichiometric redox cycles eliminate the large energy penalty associated with product separation and can be implemented in a single reactor. Without phase change, an opportunity exists for implementing solid-to-solid heat recovery to increase efficiency. Examples of materials proposed to realize the non-stoichiometric redox cycle include iron oxides (ferrites) [14], cerium dioxide (ceria) [15], [16], [17], [18], [19], [20], and mixed metal oxides such as doped ferrites [21], [22], [23] and doped ceria [16], [17], [24], [25], [26], [27], [28]. Dopants allow for significantly lower reduction temperature without lowering the melting point of the base metal oxide [21], [24]. Typical dopants are Mn, Mg, Co, Zn, Sm, and Ni. In addition, a supporting matrix structure can be added to improve oxide stability [19], [23], [29]. Ceria-based oxides are attractive for fast kinetics at both cycle steps, enabling high fuel production rates [16], [30].

The importance of heat recovery in solar thermochemical redox cycles is widely recognized (e.g., [2], [14], [21], [31]), but efforts to implement heat recovery in prototype reactors are limited. Diver et al. [14] predicts that heat recovery could improve the efficiency of an iron oxide cycle from 35% to 76%,1 and developed and conducted limited tests of a reactor that includes a system for heat recovery from the solid metal oxide. The reactor, referred to as the CR5, uses a stack of counter-rotating rings with the reactive material along the perimeter of each ring. The reactive structures serve as extended heat transfer surfaces for heat recovery [14], [21]. Each ring rotates in the opposite direction from its neighbors to permit continuous heat recovery. Tests of the CR5 in Sandia National Laboratory’s solar furnace serve as a proof of concept, but long-term continuous operation was not achieved due to a number of challenges, including thermal stressing and durability of the reactive material [32]. An alternate approach based on sequential oxidation and reduction of ferrites supported on ceramic monoliths has also been evaluated, but heat recovery from the ferrite in this arrangement requires gas phase heat transfer [19], [23], [29].

A prototype solar reactor using non-stoichiometric ceria for thermochemical dissociation of H2O and CO2 was recently demonstrated over four cycles [18]. Without solid and gas phase heat recovery, efficiency was 0.4% for both H2O and CO2 dissociation.2 The authors suggest that efficiency could be improved by decreasing the size of the aperture, increasing the thickness of insulation, and increasing the volume to surface ratio of the reactor body. While there will be some benefit to these options, the concept of increasing the insulation thickness does not recognize the important requirement to reject heat from the reduced metal oxide in order to carry out the oxidation at a lower temperature than the reduction, as required by process thermodynamics. It also ignores the potential benefits of implementing solid as well as gas phase heat recovery.

In this paper, we demonstrate the importance of both solid and gas phase heat recovery on solar-to-fuel efficiency. We analyze the effect of heat recovery from the metal oxide and from the hot product gases on the efficiency of a hypothetical dual-zone solar thermochemical reactor for the non-stoichiometric cycle (1)–((2a), (2b)). The analysis is based on the experimental redox characteristics of CeO2 available in the literature [33]. The results guide the direction of future efforts aimed at achieving solar-to-fuel efficiencies of 10% and higher.

Section snippets

Thermodynamics of non-stoichiometric CeO2

The model reacting metal oxide used in the current analysis is ceria, CeO2. Ceria is ideally suited for a reactor system with solid phase heat recovery because it has a high melting point (2800 K) and is capable of δ = 0.25 without undergoing phase change [16]. Heat recovery is simplified because the material remains in the solid phase throughout cycling. Though ongoing work seeks to improve the thermodynamic properties of CeO2 through the addition of dopants [15], [16], [25], [26], [27], [28],

Model system

The model system used to demonstrate the potential of heat recovery for increasing the efficiency of solar thermochemical fuel production using the two-step redox cycle (1)–(2) is shown in Fig. 2. The system includes a dual-zone solar thermochemical reactor for the two-step metal oxide H2O/CO2 splitting cycle with heat recovery applied to the reacting metal oxide, gaseous reactants and products, and inert gas. Reactions (1) and (2) proceed in the reduction and oxidation zones, respectively, and

Analysis

Steady-state energy balances are applied to the model system for H2O and CO2 splitting cycles. For the purpose of brevity, only the equations for the H2O splitting cycle are shown. The energy conservation equation for the entire model system shown in Fig. 2 isQ˙solarQ˙reradQ˙CeO2Q˙oxQ˙other=n˙N2[h¯N2(T2)h¯N2(T0)]+n˙O2h¯O2(T2)+n˙H2h¯H2(T4)n˙H2Oh¯H2O(l)(T0)

Note that Eq. (4) does not explicitly include the heat loss component Q˙gas because it is accounted for in the enthalpies of the gases

Cycle efficiency

In this section, we explore the potential for improving the efficiencies given by Eqs. (14), (15) by applying varying values of heat recovery to the solid and gas phases. The analysis is performed for H2O splitting using the baseline parameter set shown in Table 1, unless stated otherwise.

Here, we briefly justify the selection of the parameters for the baseline set. According to Section 2, the non-stoichiometric data are available for pO2 ≤ 0.01 atm; the upper value is used for the baseline

Conclusion

The first and second laws of thermodynamics were applied to analyze the potential of applying heat recovery for realizing high efficiency in solar-driven CeO2-based non-stoichiometric redox cycles to split H2O or CO2. Steady-state energy conservation equations were formulated for a model system by accounting for radiative exchange between the reduction zone and the surroundings, enthalpy exchange between the system and the environment by gaseous species, enthalpy flow inside the cycle with CeO2

Acknowledgments

The financial support of the National Science Foundation (grant no. EFRI-1038308) and of the Initiative for Renewable Energy and the Environment (grant nos. RL-0001-2009 and RL-0003-2011) is gratefully acknowledged.

References (38)

Cited by (180)

  • Universal expressions of solar thermochemical efficiency

    2023, Sustainable Energy Technologies and Assessments
View all citing articles on Scopus
View full text