Professional Documents
Culture Documents
Solar Energy
journal homepage: www.elsevier.com/locate/solener
George W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0405, USA
Keywords: Solar electricity production via an Air Brayton cycle is considered with integrated thermochemical energy
Concentrated solar energy storage. The storage is realized via a two-step solar thermochemical cycle based on Fe2O3/Fe3O4 reduction-
Thermochemical energy storage oxidation reactions, encompassing (1) the thermal reduction of Fe2O3 to Fe3O4 and O2 driven by concentrated
Iron oxide solar irradiation under vacuum; and (2) the exothermic oxidation of Fe3O4 with a compressed air stream back to
Thermodynamic analysis
Fe2O3. The steps may be decoupled, resulting in a high temperature, pressurized airflow that is expanded across
Kinetic analysis
a turbine to produce on-demand electricity. A thermodynamic analysis of the system determined a maximum
cycle efficiency of 46.0% at a solar concentration ratio of 4000 suns, an oxidation pressure of 30 bar, and an
approximately 5:1 molar flow rate ratio of air to solid Fe2O3 exiting the re-oxidizer. Chemical kinetics for the
thermal reduction of Fe2O3 were determined between approximately 1400 and 1700 K using non-isothermal
thermogravimetry with heating rates between 10 and 20 K·s−1 and O2 partial pressures between 0 and 0.05 bar.
The rate-limiting reaction mechanism was determined to be nucleation, and kinetic parameters were resolved
using an Avrami-Erofe’ev nucleation model with a reaction order of 1.264 ± 0.010. The rate constant followed
an Arrhenius-type temperature dependency with an apparent activation energy of 487.0 ± 3.6 kJ·mol−1 and
pre-exponential factor 2.768 ± 0.783·1014 s−1. A power-law dependence on O2 partial pressure of order
8.317 ± 0.233 was determined. Non-isothermal thermogravimetry to examine the oxidation of Fe3O4 to Fe2O3
revealed multiple kinetic regimes, and isothermal thermogravimetry showed the reaction proceeded rapidly,
within 20 s, at temperatures greater than 673 K. Solid characterization was carried out using scanning electron
microscopy and x-ray powder diffractometry up to temperatures of 1073 K to verify initial and final sample
compositions and structures.
⁎
Corresponding author.
E-mail address: peter.loutzenhiser@me.gatech.edu (P.G. Loutzenhiser).
https://doi.org/10.1016/j.solener.2018.09.043
Received 14 July 2018; Received in revised form 14 September 2018; Accepted 16 September 2018
Available online 24 September 2018
0038-092X/ © 2018 Elsevier Ltd. All rights reserved.
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Subscripts
∞ equilibrium value
Imponenti et al., 2017; Zhang et al., 2016). Fe2O3/Fe3O4 redox reac- metals, with the goal of lowering the thermal reduction temperature
tions are particularly attractive because Fe2O3 has a higher thermal (André et al., 2017; Block et al., 2014; Block and Schmücker, 2016;
reduction temperature than most other binary candidates, permitting Pagkoura et al., 2014). However, past efforts have shown a significant
higher theoretical Air Brayton cycle efficiencies. The Fe2O3/Fe3O4 decrease in reaction enthalpy.
redox pair materials are also are relatively inexpensive and widely In this study, the Fe2O3 thermal reduction temperature was reduced
available compared to Co3O4/CoO materials, and they carry fewer en- by operating at decreased O2 partial pressures to shift the equilibrium
vironmental/human health concerns (Smith and Huyck, 1999), making to a more favorable temperature range according to Le Chatelier’s
the redox pair a more practical and economically appealing candidate principle. The impact of decreasing the O2 partial pressure is shown in
for large-scale deployment. The reversible redox reaction of Fe2O3/ Fig. 1. For an O2 partial pressure of 10−3 bar, Fe2O3 begins to reduce to
Fe3O4 is represented as: Fe3O4 at a temperature of 1432 K. The Fe2O3/Fe3O4 pair allows for
higher oxidation temperatures and, therefore, greater Air Brayton cycle
1
3Fe2 O3 2Fe3O4 + O2(g) , H298.15 K = 78.3 kJ·mol 1
theoretical efficiencies, at the cost of decreasing receiver absorption
2 (1)
efficiencies due to greater re-radiative losses to the environment
Iron oxides have been the focus of numerous solar thermochemical (Steinfeld and Palumbo, 2001). Therefore, a comprehensive thermo-
studies, including carbothermal iron production (Steinfeld and Fletcher, dynamic cycle analysis was performed to assess the potential of the
1991), fuels co-production via carbothermal reduction and/or methane Fe2O3/Fe3O4 pair.
cracking (Halmann and Steinfeld, 2006; Steinfeld et al., 1995; Steinfeld An analysis of Fe2O3/Fe3O4 TCES integrated into an Air Brayton
et al., 1993; Tamaura et al., 1997), and H2O and/or CO2 splitting via cycle was performed as a function of molar flow rate of air, reactor
two-step cycles based on Fe3O4/FeO redox reactions (Galvez et al., temperature, compressor outlet/turbine inlet pressure, and solar con-
2008; Loutzenhiser et al., 2009; Nakamura, 1977; Sibieude et al., 1982; centration ratio. The thermal reduction and oxidation reactions were
Stamatiou et al., 2010). For the Fe2O3/Fe3O4 pair, the high Fe2O3 examined using thermogravimetric analysis for a range of heating rates
thermal reduction temperatures have presented both materials com- and O2 partial pressures to measure reaction rates, identify rate limiting
patibility and engineering design challenges. Such concerns have mechanism(s), and determine kinetic parameters. A cycling study was
prompted the study of combined oxides of iron and other multivalent performed to examine material stability and scanning electron
618
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Fig. 2. System schematic of the Air Brayton cycle with an integrated two-step thermochemical energy storage cycle based on Fe2O3/Fe3O4 redox reactions with
relevant mass and energy flows.
619
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Table 1
Model property values for the analysis of cycle efficiency vs. solar concentration
ratio for the Air Brayton cycle with paired thermochemical energy storage using
Fe2O3/Fe3O4 and Co3O4/CoO reduction-oxidation pairs.
Property Fe2O3/Fe3O4 Co3O4/CoO
Fig. 3. Fe2O3/Fe3O4 particle temperature exiting the re-oxidizer (solid) and O2-
depleted air temperature entering the turbine (dashed), as functions of molar
air flow rate and for a solar thermochemical reactor temperature of 1432 K,
solar concentration ratio of 1000 suns, and re-oxidizer pressure of 30 bar.
with the abrupt Trecover minimum in Fig. 3, the required input reactor
heat transfer rate Qsolar reached a maximum for nair=5.1 mol·s−1, as the
particles were fully re-oxidized at the minimum allowable temperature,
providing no more chemical or sensible energy storage potential for
higher nair . The turbine power Wturbine = increased slower for
nair >5.1 mol·s−1 but did not become zero, as the increasing nair con-
Fig. 5. Theoretical cycle efficiencies for Fe2O3/Fe3O4 (solid) and Co3O4/CoO
tinued to increase the power output.
(dashed) thermochemical cycles as a function of solar concentration ratio at
The maximum theoretical thermodynamic cycle efficiency for the receiver temperatures of 1432 and 1037 K and compressor airflow rates of 5.1
Air Brayton-integrated thermochemical cycle was defined as: and 13.6 mol·s−1, respectively, and a re-oxidizer pressure of 30 bar.
Wturbine Wcomp Wvac
=
(2)
cycle
Qsolar
Table 1.
At C = 1000 suns, Treactor = 1432 K, pcomp = 30 bar, and The ηcycle for Fe2O3/Fe3O4 increased more rapidly with C than for
nair = 5.1 mol·s−1, a ηcycle = 38.4% was determined, which is sig- Co3O4/CoO due to a higher Treactor, as shown in Fig. 5, and the ηcycle for
nificantly lower than the ηcycle = 43.4% from the Co3O4/CoO analysis the Fe2O3/Fe3O4 TCES cycle exceeded the Co3O4/CoO cycle at
under equivalent conditions (Schrader et al., 2015). The lower ηcycle C ≥ 4352 suns.
was primarily due to significantly higher Treactor and, therefore, higher Fig. 6 shows nair for a range of pcomp and Treactor. A ηcycle = 46.0%
Qloss for the Fe2O3/Fe3O4 pair. However, because Qloss is mitigated by was achieved for C = 4000 suns, Treactor = 1432 K, pcomp = 30 bar, and
increasing C (Steinfeld et al., 1998; Steinfeld and Palumbo, 2001), an nair = 5.1 mol·s−1, and the ηcycle increased with diminishing returns for
analysis of both redox pairs was performed to determine how ηcycle greater nair and C. Increasing pcomp increased ηcycle as the O2-deficient
varies with C for Treactor = TΔG=0 and nair fixed to achieve full re-oxi- airflow entering the turbine expanded across a greater temperature
dation and maximum sensible energy recovery in the re-oxidizer. Re- difference, while increasing Treactor decreased ηcycle due to rapid in-
levant model property value inputs for the analysis are shown in creases in Qloss from reradiation.
620
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Fig. 6. Theoretical cycle efficiency as a function of molar air flow rate entering the re-oxidizer for (a) compressor pressures of 5, 17.5, and 30 bar and a solar
thermochemical reactor temperature of 1432 K; and (b) solar thermochemical reactor temperatures of 1432, 1532, and 1632 K and a compressor pressure of 30 bar.
3. Kinetic analyses confidence intervals for the parameters p1 … pj using the 2.5th and
97.5th percentile values of each fitted parameter (Alper and Gelb,
The thermal reduction of Fe2O3 to Fe3O4 and O2 and re-oxidation of 1990).
Fe3O4 to Fe2O3 were examined in a thermogravimetric analyzer (TGA,
Netzsch STA 449 F3 Jupiter, ± 1 μg). Separate unreacted, high-purity
Fe2O3 and Fe3O4 samples were selected for the two reaction steps to 3.1. Fe2O3 thermal reduction
ensure that samples were not initially in partially reduced or oxidized
states, respectively. Powder samples were placed on an Al2O3 crucible The rate limiting mechanism of the Fe2O3 thermal reduction was
shielded with a platinum foil (Sigma-Aldrich 0.025MM Thick 99% Pt described with a variable-order Avrami-Erofe’ev nucleation-controlled
Foil 267244-1.4G). Sample temperatures were measured by an S-type model (AN), given as:
thermocouple. Sample masses were measured before and after testing
n-1
with an analytical balance (Mettler-Toledo ML54, ± 0.1 mg) to verify f ( ) = n (1 )[ ln(1 )] n (7)
the total mass change measured by TGA. Blank runs under identical
conditions were performed following each experiment to correct for the where n is the reaction order. k0, Ea, and n were simultaneously fitted at
influences of buoyancy and gas dynamics. multiple heating rates to verify that the reaction was not heat transfer
The temporal conversion from Fe2O3 to Fe3O4 and vice versa were limited (Vyazovkin et al., 2011). The three fitted values were then fixed
calculated as: and h (pO2 ) was described using a power-law relationship, given as:
m (t ) m 0 pO2
m
(t ) =
m m0 (3) h (pO2 ) = 1
pO2, eq (T ) (8)
where m(t) is the mass at time t; m0 is the initial sample mass, between
90 and 100 mg; and m∞ is the stoichiometric equilibrium mass for the where pO2,eq is the equilibrium partial pressure for T and m is the
fully-reacted sample. The reaction rate was defined as: constant reaction order for the pO2 dependence, which was simulta-
neously fitted to experimental runs performed at multiple pO2 .
d
r= = h (pO2 ) k (T ) f ( ) All non-isothermal reduction experiments were performed with a
dt (4)
constant total gas flow of 100 mLN·min−1 (where LN refers to liters at
where the temporal conversion rate dα/dt was calculated with a fourth- standard conditions, i.e., 273 K and 1 bar). Fe2O3 samples (Sigma-
order finite difference approximation; and a reaction rate constant k(T), Aldrich ≥99.995 trace metals basis, particle size < 5 μm) were heated
partial pressure dependence h (pO2 ) , and conversion-dependent kinetic to 1723 K in Ar at heating rates of 10–20 K·min−1 in the TGA high
model f(α) were fitted to r using a Nelder-Mead nonlinear minimization temperature graphite furnace. The thermal reduction proceeded at
algorithm with a standard squared error objective function, given as: elevated temperatures, and samples agglomerated and turned dark
gray, similar to unreacted Fe3O4.
SSE = [ri ri (t , T, , pO2 ; p1, p2, …, pj)]2 The relationship between α and T was found to be independent of β,
i (5)
indicating no heating rate (heat transfer) limitations. The reaction rate
where a least squares criterion was used to determine r , the modeled consistently slowed significantly for values α > 0.90, and an iso-
reaction rate; t, T, α, and pO2 were known for the datapoints i; and the conversional analysis (Friedman, 1964) revealed large changes in Ea for
unknown parameters p1 … pj were simultaneously fitted. An Arrhenius- such values. Thus, the analysis was limited to 0 < α < 0.90, and
type temperature dependency was assumed, given as: k0 = 2.768 ± 0.783 · 1014 s−1, Ea = 487.0 ± 3.6 kJ·mol−1, and
n = 1.264 ± 0.010 were simultaneously fitted with 95% confidence
Ea
k (T ) = k 0 exp intervals from the Monte Carlo analysis.
RT (6)
Two methodologies were employed to validate the model-fitted
where k0 is the apparent pre-exponential factor; Ea is the apparent ac- parameters. To evaluate the nucleation model, reaction order, and the
tivation energy; and R is the universal gas constant. A Monte Carlo error appropriateness of the Arrhenius form of k, an Arrhenius plot was
analysis using 500 iterations was performed to determine 95% produced by calculating k as:
621
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
622
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Fig. 10. Schematic of Fe3O4 oxidation pathways (a) at the onset of the diffusion-limited regime, resulting in (b) oxidation to maghemite, or γ-Fe2O3; and (c) at the
transition to the nucleation-limited regime, resulting in (d) oxidation to hematite, or α-Fe2O3.
623
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Fig. 12. Experimental conversion for isothermal Fe3O4 oxidation at an O2 concentration of 80% and temperatures between 673 and 973 K: (a) the full oxidation
profiles and (b) detail-views of the first 25 s of oxidation.
environment, then cycled between 0% and 20% O2–Ar in 15 min in- to Fig. 14c, the oxidation rates did not slow with cycle number, and in
tervals at a flow rate of 100 mLN·min−1. T = 1623 K was selected to fact the oxidation of cycles 2–9 was consistently more rapid than cycle
allow for the thermodynamic favorability of both cycle steps for the 1, as the majority of αn values were above the identity line.
given change in pO2 , and to achieve rapid reaction kinetics. The relative
temporal mass change Δm/m0 is shown in Fig. 13, and a dotted line 5. Solid characterization
indicates the theoretical Δm/m0 for full conversion. Including an initial
bakeout (cycle 0), the sample underwent 10 redox cycles and a cool- X-ray diffractometry (XRD, PANalytical X’Pert PRO Alpha-1 dif-
down in an oxidizing environment, shown in gray. The sample did not fractometer) was performed on reacted and unreacted samples of Fe2O3
fully re-oxidize during the bakeout cycle nor in any cycle after, con- and Fe3O4 at diffraction angles 2θ of 20–70° to verify sample compo-
sistent with thermodynamic equilibrium calculations (ΔG = 0) which sitions. Fig. 15 shows the intensity peaks for both species, with the
forecast a Fe3O4–Fe2O3 mixture for pO2 = 0.2 bar and T > 1100 K. unreacted (a) Fe3O4 and (b) Fe2O3 sample peaks shifted up for com-
While there was some variation in the oxidation extents per cycle due to parison with the corresponding (a) thermally reduced and (b) oxidized
slowed reaction rates as the sample neared thermodynamic equili- samples, respectively. The unreacted sample crystalline structures were
brium, there was no apparent systematic change in redox capacity first verified via comparison to star-quality XRD peaks from the PDF4+
during cycles 1–9. database (ICDD, 2016). Reacted samples were then compared to their
Fig. 14 depicts (a) the relative temporal sample mass changes for unreacted form to confirm complete thermal reduction and/or oxida-
cycles 0, 1, 5, and 9 superimposed for comparison and parity plots tion. Good agreement between sample peak locations and relative in-
showing α from cycles 3, 5, 7, and 9 as a function of α from cycle 1 for tensities was observed for both oxidation states with their unreacted
(b) thermal reduction and (c) oxidation, respectively. Both Fe2O3 counterparts, without apparent 2θ shifting or evidence of intermediate
thermal reduction and initial Fe3O4 oxidation proceeded rapidly, con- phases.
sistent with the kinetic analyses. During the thermal reduction stage, In-situ XRD (PANalytical Empyrean with Anton-Paar 1200N Hot
full conversion was achieved, supporting the assumption of complete Stage) was also performed to examine changes in the crystalline
thermal reduction in the solar reactor for the thermodynamic analysis. structures of oxidized Fe3O4 samples at elevated temperatures. Non-
For the re-oxidation stage, the final conversion varied significantly from reacting samples of Fe2O3 were heated to 1073 K in an atmospheric
cycle to cycle, with a maximum near α = 0.8. However, the sample did environment with diffraction peaks obtained in intervals of 200 K. The
approach α = 1 during the oxidizing cooldown, as shown in the gray resulting structures were compared to PDF4+ database entries for both
region (t > 18000 s) of Fig. 13, as the full oxidation of Fe3O4 to Fe2O3 -Fe2O3 and -Fe2O3 and are shown in Fig. 16. While there was some
became thermodynamically favorable at lower temperatures. According evidence of peak shifting, particularly at higher 2θ peaks, to lower
624
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Fig. 14. For initially-Fe2O3 samples, (a) superimposed reduction-oxidation cycle mass changes relative to initial sample mass; parity plots for (b) thermal reduction
and (c) re-oxidation comparing conversion to the first cycle.
Fig. 15. Intensity as a function of 2θ angle from X-ray diffractometry measurements for (a) an unreacted Fe3O4 sample and Fe3O4 produced via thermal reduction of
Fe2O3, (b) an unreacted Fe2O3 sample and Fe2O3 produced from oxidation of Fe3O4.
used to study particle size and morphology for the Fe2O3 and Fe3O4
samples (a) before and (b) after thermal reduction, shown in Fig. 17,
and oxidation, shown in Fig. 18, respectively. The majority of unreacted
Fe2O3 particles were much smaller than the 5 μm maximum specified
diameter, with smooth surfaces and near spherical shapes. Following
thermal reduction, the particle sizes grew significantly, resulting in less
spherical, slightly porous particles. The unreacted Fe3O4 particles had a
relatively wide range of particle sizes despite sieving, as well as more
angular initial surface shapes. Oxidation resulted in a porous, angular
particle sample in which the smaller particles were no longer present.
However, the apparent particle size growth was less significant relative
to the initially-Fe2O3 sample.
6. Conclusions
625
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Fig. 17. Scanning electron microscopy images for (a) an unreacted Fe2O3 sample and (b) Fe3O4 produced via thermal reduction of Fe2O3.
Fig. 18. Scanning electron microscopy images for (a) an unreacted Fe3O4 sample and (b) Fe2O3 produced via oxidation of Fe3O4.
revealed that significant oxidation within 20 s may be achieved. thermal radiative flux. Energy Fuels 27 (8), 4884–4890.
Pressure-swing Fe2O3/Fe3O4 cycling showed reduced conversion fol- Alonso, E., Pérez-Rábago, C., Licurgo, J., Fuentealba, E., Estrada, C.A., 2015. First ex-
perimental studies of solar redox reactions of copper oxides for thermochemical
lowing a bakeout thermal reduction due to thermodynamic limitations, energy storage. Sol. Energy 115, 297–305.
but the redox capacity did not systematically change over the following Alper, J.S., Gelb, R.I., 1990. Standard errors and confidence intervals in nonlinear re-
nine cycles. SEM revealed significant Fe2O3 particle size growth and an gression: comparison of Monte Carlo and parametric statistics. J. Phys. Chem. 94
(11), 4747–4751.
apparent decrease in surface area during thermal reduction. André, L., Abanades, S., Cassayre, L., 2017. High-temperature thermochemical energy
The combined thermodynamic-kinetic analysis of a solar thermo- storage based on redox reactions using Co-Fe and Mn-Fe mixed metal oxides. J. Solid
chemical cycle based on Fe2O3/Fe3O4 redox reactions coupled to an Air State Chem. 253 (Suppl. C), 6–14.
Babiniec, S.M., Coker, E.N., Miller, J.E., Ambrosini, A., 2015. Investigation of
Brayton cycle indicated superior maximum theoretical efficiencies for LaxSr1−xCoyM1−yO3−δ (M = Mn, Fe) perovskite materials as thermochemical energy
concentrating solar infrastructure capable of very high solar con- storage media. Sol. Energy 118, 451–459.
centration ratios. Kinetic analyses and cycling demonstrated rapid Babiniec, S.M., Coker, E.N., Miller, J.E., Ambrosini, A., 2016. Doped calcium manganites
for advanced high-temperature thermochemical energy storage. Int. J. Energy Res. 40
conversion at representative temperature and pressure conditions for
(2), 280–284.
both reactions. This work is an important prerequisite for guiding the Becker, K.D., von Wurmb, V., Litterst, F.J., 1993. Disorder and diffusional motion of
design and optimization of solar thermochemical reactors and non-solar cations in magnetite, Fe3−δO4, studied by Mössbauer spectroscopy at high tem-
re-oxidizers. These results establish an important upper bound on the peratures. J. Phys. Chem. Solids 54 (8), 923–935.
Block, T., Knoblauch, N., Schmücker, M., 2014. The cobalt-oxide/iron-oxide binary
performance of future implementations and inform the design and op- system for use as high temperature thermochemical energy storage material.
timization of thermochemical infrastructure. Thermochimica Acta 577 (Suppl. C), 25–32.
Block, T., Schmücker, M., 2016. Metal oxides for thermochemical energy storage: a
comparison of several metal oxide systems. Sol. Energy 126 (Suppl. C), 195–207.
Acknowledgements Bowrey, R.G., Jutsen, J., 1978. Energy storage using the reversible oxidation of barium
oxide. Sol. Energy 21 (6), 523–525.
This material is based upon work supported by the National Science Brendelberger, S., von Storch, H., Bulfin, B., Sattler, C., 2017. Vacuum pumping options
for application in solar thermochemical redox cycles – assessment of mechanical-, jet-
Foundation Graduate Research Fellowship under Grant No. DGE- and thermochemical pumping systems. Sol. Energy 141, 91–102.
1650044. Any opinions, findings, and conclusions or recommendations Carrillo, A.J., Sastre, D., Serrano, D.P., Pizarro, P., Coronado, J.M., 2016. Revisiting the
expressed in this material are those of the author(s) and do not ne- BaO2/BaO redox cycle for solar thermochemical energy storage. PCCP 18 (11),
8039–8048.
cessarily reflect the views of the National Science Foundation. The
Carrillo, A.J., Serrano, D.P., Pizarro, P., Coronado, J.M., 2014. Thermochemical heat
authors wish to thank Andrew Schrader for access to and discussion of storage based on the Mn2O3/Mn3O4 redox couple: influence of the initial particle size
his computational models for adaptation in this work. on the morphological evolution and cyclability. J. Mater. Chem. A 2 (45),
19435–19443.
Deutsch, M., Horvath, F., Knoll, C., Lager, D., Gierl-Mayer, C., Weinberger, P., Winter, F.,
References 2017. High-temperature energy storage: kinetic investigations of the CuO/Cu2O re-
action cycle. Energy Fuels 31 (3), 2324–2334.
Agrafiotis, C., Roeb, M., Schmuecker, M., Sattler, C., 2014. Exploitation of thermo- Feitknecht, W., Gallagher, K.J., 1970. Mechanisms for the oxidation of Fe3O4. Nature 228,
chemical cycles based on solid oxide redox systems for thermochemical storage of 548.
solar heat. Part 1: Testing of cobalt oxide-based powders. Sol. Energy 102, 189–211. Friedman, H.L., 1964. Kinetics of thermal degradation of char-forming plastics from
Alonso, E., Hutter, C., Romero, M., Steinfeld, A., Gonzalez-Aguilar, J., 2013. Kinetics of thermogravimetry. Application to a phenolic plastic. J. Polym. Sci. Part C: Polym.
Mn2O3–Mn3O4 and Mn3O4–MnO redox reactions performed under concentrated Symp. 6 (1), 183–195.
626
H.E. Bush, P.G. Loutzenhiser Solar Energy 174 (2018) 617–627
Gallagher, K.J., Feitknecht, W., Mannweiler, U., 1968. Mechanism of oxidation of mag- and modeling. Sol. Energy 150, 584–595.
netite to γ-Fe2O3. Nature 217, 1118. Schrader, A.J., Muroyama, A.P., Loutzenhiser, P.G., 2015. Solar electricity via an Air
Galvez, M., Loutzenhiser, P., Hischier, I., Steinfeld, A., 2008. CO2 splitting via two-step Brayton cycle with an integrated two-step thermochemical cycle for heat storage
solar thermochemical cycles with Zn/ZnO and FeO/Fe3O4 redox reactions: thermo- based on Co3O4/CoO redox reactions: thermodynamic analysis. Sol. Energy 118,
dynamic analysis. Energy Fuels 22 (5), 3544–3550. 485–495.
Gorman, B.T., Johnson, N.G., Miller, J.E., Stechel, E.B., 2015. Thermodynamic Shebanova, O.N., Lazor, P., 2003. Raman study of magnetite (Fe3O4): LASER-induced
Investigation of Concentrating Solar Power With Thermochemical Storage. (56840), thermal effects and oxidation. J. Raman Spectrosc. 34 (11), 845–852.
V001T005A026. Sibieude, F., Ducarroir, M., Tofighi, A., Ambriz, J., 1982. High temperature experiments
Halmann, M., Steinfeld, A., 2006. Fuel saving, carbon dioxide emission avoidance, and with a solar furnace: the decomposition of Fe3O4, Mn3O4, CdO. Int. J. Hydrogen
syngas production by tri-reforming of flue gases from coal- and gas-fired power Energy 7 (1), 79–88.
stations, and by the carbothermic reduction of iron oxide. Energy 31 (15), Sidhu, P.S., Gilkes, R.J., Posner, A.M., 1977. Mechanism of the low temperature oxidation
3171–3185. of synthetic magnetites. J. Inorg. Nucl. Chem. 39 (11), 1953–1958.
Haseli, P., Jafarian, M., Nathan, G.J., 2017. High temperature solar thermochemical SIGMA-ALDRICH, 2017. Safety Data Sheet: cobalt(II) oxide [1308-06-1]. Sigma-Aldrich,
process for production of stored energy and oxygen based on CuO/Cu2O redox re- Inc, Darmstadt, Germany.
actions. Sol. Energy 153, 1–10. Sahoo, S.K., Agarwal, K., Singh, A.K., Polke, B.G., Raha, K.C., 2010. Characterization of γ-
Hutchings, K.N., Wilson, M., Larsen, P.A., Cutler, R.A., 2006. Kinetic and thermodynamic and α-Fe2O3 nano powders synthesized by emulsion precipitation-calcination route
considerations for oxygen absorption/desorption using cobalt oxide. Solid State Ion. and rheological behaviour of α-Fe2O3. Int. J. Eng. Sci. Technol. 2 (8), 118–126.
177 (1), 45–51. Smith, K.S., Huyck, H.L., 1999. An overview of the abundance, relative mobility, bioa-
ICDD, 2016. PDF-4+ 2016. International Centre for Diffraction Data, Newtown Square, vailability, and human toxicity of metals. The Environ. Geochem. Mineral Deposits 6,
PA, USA. 29–70.
Imponenti, L., Albrecht, K.J., Wands, J.W., Sanders, M.D., Jackson, G.S., 2017. Stamatiou, A., Loutzenhiser, P.G., Steinfeld, A., 2010. Solar syngas production via H2O/
Thermochemical energy storage in strontium-doped calcium manganites for con- CO2-splitting thermochemical cycles with Zn/ZnO and FeO/Fe3O4 redox reactions.
centrating solar power applications. Sol. Energy 151, 1–13. Chem. Mater. 22 (3), 851–859.
Kissinger, H.E., 1957. Reaction kinetics in differential thermal analysis. Anal. Chem. 29 Steinfeld, A., Fletcher, E.A., 1991. Theoretical and experimental investigation of the
(11), 1702–1706. carbothermic reduction of Fe2O3 using solar-energy. Energy 16 (7), 1011–1019.
Linstrom, P.J., Mallard, W.G., 2001. The NIST Chemistry WebBook: a chemical data re- Steinfeld, A., Frei, A., Kuhn, P., 1995. Thermoanalysis of the combined Fe3O4-reduction
source on the internet. J. Chem. Eng. Data 46 (5), 1059–1063. and CH4-reforming processes. Metall. Mater. Trans. B 26 (3), 509–515.
Loutzenhiser, P.G., Gálvez, M.E., Hischier, I., Stamatiou, A., Frei, A., Steinfeld, A., 2009. Steinfeld, A., Kuhn, P., Karni, J., 1993. High-temperature solar thermochemistry: pro-
CO2 splitting via two-step solar thermochemical cycles with Zn/ZnO and FeO/Fe3O4 duction of iron and synthesis gas by Fe3O4-reduction with methane. Energy 18 (3),
redox reactions II: kinetic analysis. Energy Fuels 23 (5), 2832–2839. 239–249.
Monazam, E.R., Breault, R.W., Siriwardane, R., 2014. Kinetics of magnetite (Fe3O4) Steinfeld, A., Kuhn, P., Reller, A., Palumbo, R., Murray, J., Tamaura, Y., 1998. Solar-
oxidation to hematite (Fe2O3) in air for chemical looping combustion. Ind. Eng. processed metals as clean energy carriers and water-splitters. Int. J. Hydrogen Energy
Chem. Res. 53 (34), 13320–13328. 23 (9), 767–774.
Muroyama, A.P., Schrader, A.J., Loutzenhiser, P.G., 2015. Solar electricity via an Air Steinfeld, A., Palumbo, R., 2001. Solar thermochemical process technology. Encyclopedia
Brayton cycle with an integrated two-step thermochemical cycle for heat storage Phys. Sci. Technol. 15 (1), 237–256.
based on Co3O4/CoO redox reactions II: Kinetic analyses. Sol. Energy 122, 409–418. Tamaura, T., Tsuji, M., Ehrensberger, K., Steinfeld, A., 1997. The solar-driven coal/Fe3O4
Nakamura, T., 1977. Hydrogen production from water utilizing solar heat at high tem- redox system. J. Phys. IV France 07 (C1) C1-685-C681-686.
peratures. Sol. Energy 19 (5), 467–475. Tiernan, M.J., Barnes, P.A., Parkes, G.M.B., 2001. Reduction of iron oxide catalysts: the
Neises, M., Tescari, S., de Oliveira, L., Roeb, M., Sattler, C., Wong, B., 2012. Solar-heated investigation of kinetic parameters using rate perturbation and linear heating ther-
rotary kiln for thermochemical energy storage. Sol. Energy 86 (10), 3040–3048. moanalytical techniques. J. Phys. Chem. B 105 (1), 220–228.
Pagkoura, C., Karagiannakis, G., Zygogianni, A., Lorentzou, S., Kostoglou, M., Vyazovkin, S., Burnham, A.K., Criado, J.M., Pérez-Maqueda, L.A., Popescu, C.,
Konstandopoulos, A.G., Rattenburry, M., Woodhead, J.W., 2014. Cobalt oxide based Sbirrazzuoli, N., 2011. ICTAC Kinetics Committee recommendations for performing
structured bodies as redox thermochemical heat storage medium for future CSP kinetic computations on thermal analysis data. Thermochim. Acta 520 (1), 1–19.
plants. Sol. Energy 108, 146–163. Wong, B., 2011. Thermochemical heat storage for concentrated solar power, final report
Papanastassiou, D., Bitsianes, G., 1973. Mechanisms and kinetics underlying the oxida- for the US Department of Energy. San Diego, CA, USA.
tion of magnetite in the induration of iron ore pellets. Metall. Trans. 4 (2), 487–496. Yu, J., Han, Y., Li, Y., Gao, P., Li, W., 2017. Mechanism and kinetics of the reduction of
Parkinson, G.S., 2016. Iron oxide surfaces. Surf. Sci. Rep. 71 (1), 272–365. hematite to magnetite with CO–CO2 in a micro-fluidized bed. Minerals 7 (11), 209.
Schrader, A.J., De Dominicis, G., Schieber, G.L., Loutzenhiser, P.G., 2017. Solar electricity Zhang, Z., Andre, L., Abanades, S., 2016. Experimental assessment of oxygen exchange
via an Air Brayton cycle with an integrated two-step thermochemical cycle for heat capacity and thermochemical redox cycle behavior of Ba and Sr series perovskites for
storage based on Co3O4/CoO redox reactions III: Solar thermochemical reactor design solar energy storage. Sol. Energy 134, 494–502.
627