CC BY
42SOLUTION COMBUSTION SYNTHESIS OF DOPED Ca2AlMnO5 AS A NEW OXYGEN STORAGE MATERIAL
A. Sato, G. Saito, K. Abe, T. Nomura, and T. Akiyama
Center for Advanced Research of Energy and Materials, Hokkaido University, Kita 13 Nishi 8 kita-ku, Sapporo, 060-8628 Japan *e-mail: [email protected]
DOI: 10.24411/9999-0014A-2019-10149
Oxygen is used in many fields such as steelmaking, garbage incinerator, welding, rocket fuel, etc. Although cryogenic separation from air is a mainstream of the current oxygen production method, PSA (Pressure Swing Adsorption) process using an oxygen storage material (OSM) as a more energy saving method attracts attention. As the OSM, Ca2AlMnO5 (CAMO) is a promising because oxygen storage/release performance is high [1]. As shown in Fig.1, CAMO has a brownmillerite-type structure whose chemical composition is expressed as A2B2O5 in which oxygen vacancies of the perovskite structure ABO3 are ordered [2]. The B site has a composition in which Al and Mn occupy orderly at a ratio of 1:1. AlO4 tetrahedral layer and MnO6 octahedral layer are alternately arranged, but when oxygen is occluded, the AlO4 tetrahedron partially oxidizes and changes to AlO6 octahedron. The following reaction occurs in the CAMO during oxygen adsorption/desorption:
b
(a) (b)
Fig. 1. Schematic illustration of the crystal structures of (a) Brownmillerite-type Ca2AlMnO5 and (b) its fully oxygenated form Ca2AlMnO5+s [1].
Ca2AlMnO5 + - O2
Ca2AlMnO5+s
(1)
The above reversible absorption/desorption reaction is induced by temperature and pressure changes. However, CAMO has problems such as lowering the operating temperature range, reducing hysteresis, increasing the speed of adsorption and desorption. Improvement of properties can be expected by doping Ca and Mn sites in CAMO with trace elements, but it is still in the theoretical stage and not verified experimentally [3]. Therefore, in order to improve the properties, we attempted to dope La into the Ca site of CAMO. Doping of trivalent La into bivalent Ca will cause the lattice size change due to the change in the valence of Mn and the difference of the ion radii between La and Ca (1.12 and 1.17 A, respectively). In this study, we investigated the effects of La doping into Ca sites in CAMO on the oxygen storage characteristics.
We tried to synthesize CAMO by solution combustion synthesis. Solution combustion synthesis (SCS) is a combustion synthesis process based on a highly exothermic, selfsustaining reaction generated by heating a solution mixture of aqueous metal salts and fuels such as urea, citric acid, and glycine. This approach has been used to synthesize a variety of compounds including binary and complex oxides, such as ferrites, spinels, and perovskites [4-6]. This
c
method not only yields nanosized products exhibiting large specific surface areas but also enables uniform (homogeneous) doping of trace amounts of various elements in a single step. The characteristics (including the purity, structure and size) of the SCS oxide powders are typically determined by several synthetic parameters, such as the species of fuel and oxidizer reactants, the fuel-to-oxidizer ratio, and the subsequent sintering treatment after SCS [7]. In addition, in this experiment by SCS, the synthesis time is also very short, about 1 min.
Precursors of LaxCa(2-x)AlMnO5 were prepared by a solution combustion synthesis (SCS) method. Reagents of Ca(NOs)2 • 4H2O (99.0%), Al(NOs)3 • 9H2O (99.9%), Mn(NOs)2 (50.0% aqueous solution), La(NO3)3 • 6H2O (99.9%) and glycine (99.0%) were used as raw materials. Glycine was charged as a fuel for the SCS. As an index for determining the amount of glycine to be added, the fuel ratio O was defined by the following formula.
In this reaction the relationship between <P and n is <P = n/5. Here, <P =1 means a stoichiometric sample in Eq. (2), <P <1 means a sample with a small amount of fuel (= reducing agent), and <P >1 means a fuel excess condition. The fuel ratio of all samples was set to <P = 1 [8]. The SCS reaction of LaxCa(2-x)AlMnO5 is expressed by the following equation.
(2 - x)Ca(NO3)2 • 4H2O + Al(NO3)3 • 9H2O + Mn(NO3)2 + xLa(NO3> • 6H2O + (3) nC2H5O2N + XO2 ^ LaxCa(2-x)AlMnO5 + Gases
The reagents were weighed according to the eq. (3) and 10 ml of distilled water was added to them. They were stirred and dried at 90°C for 4 h to obtain a gel-like sample. This gel-like sample was put in a cylindrical reaction vessel preheated to 400°C to carry out the SCS. A schematic diagram of the SCS equipment used in this experiment is shown in Fig. 2.
Fig. 2. Schematic illustration of the solution combustion synthesis.
The obtained sample was annealed at 1250°C in air for 12 h and then annealed again at the same temperature and time under an Ar atmosphere to remove excess oxygen in the sample. Phase identification of the obtained product was conducted by XRD, and oxygen adsorption/desorption of the SCS samples was examined by TG-DSC.
Figure 3 shows the XRD patterns of the SCSed LaxCa(2-x)AlMnO5. The main peaks of all the SCSed samples corresponded to those of the brown mirror light structure. That meant substitution of La to the Ca-site maintained the CAMO structure. The peaks of perovskite-type impurity phases were observed when La was excessively doped (x ^ 0.1). In addition, as the amount of La added increases, the peak of CAMO weakens, and the peak of the perovskite-type impurity phase appears more notably. The simple perovskite phase other than the CAMO phase occurred at x ^ 0.1 and did not occur at x ^ 0.05.
<P =
total valency of fuel
(2)
total valency of oxidizer
LaxCa2_xAlMnO5
O: Ca2AlMnO5 A: Perovskite impurity phase
s
.1? VI
C
u
s
0 x=0 1 0 0 0 L .
0 x=0.001 1 0 S 0 0 L. .
° x=0.01 o op ib o o !..
0 x=0.02 x=0.02 0 o OIO ill 0 0 L .
0 x=0.05 1 o o9o ill o o L. .
I x=ft1 0 o J- o o
0 x=0.3 AO O A O AO
Ni m
0 x=0.5 AO o a|o jV A O AO
10
20
30
29 / deg
40
50
Fig. 3. XRD patterns of LaxCa(2-x)AlMnO5.
3.5
o
es &
O <D
SÎ —
O
Ö
<D M
X O
2.5
1.5
0.5
r LaxCa(2.x)AlMnO5
o
o
o
0 0.2 0.4 0.6
x, Concentration of La / mol
Fig. 4. Relationship between the amount of La doped and the oxygen storage amount
4
3
2
0
Figure 4 shows the maximum oxygen storage amounts of the La-doped CAMO. The oxygen storage amounts were evaluated based on the thermogravimetric change of the samples. The amount of stored oxygen was slightly increased up to x ^ 0.05. On the other hand, when x ^ 0.1, the storage amount of oxygen was greatly reduced with the addition amount of La. The decrease in the oxygen storage amount was due to the perovskite impurity phase. Figure 5 shows the relationship between the amount of La doped and the lattice volume of the SCSed CAMO. The lattice volume was calculated based on the XRD data. There was two different tendencies in the change of the lattice volume; the lattice volume increased with larger amount of doped La at x < 0.10 but decreased at x ^ 0.10. At x < 0.10, the doped La increased the lattice constant; the ion radii of La and Ca are 1.7 and 1.12 A, respectively. On the other hand, the lattice shrank with the larger amount of doped La at the range of x ^ 0.1. A perovskite impurity phase was generated at x ^ 0.1, resulting in the decrease in lattice constant.
0.435
ö
<D
J 0.43
o
>
<D O
£ cd J
0.425
0
0.1 0.2 0.3 0.4 0.5 0.6 x, Concentration of La / mol Fig. 5. Relationship between the amount of La doped and lattice volume.
0.424 0.426 0.428 0.43 0.432 0.434 Lattice volume / nm3
Fig. 6. Relationship between the lattice volume and the oxygen storage amount.
La-doped CAMO was successfully synthesized by SCS. Although the oxygen storage capacity decreased when La doped was too much, the oxygen storage capacity increased with an appropriate amount of La doped. When it was La doped too much, a perovskite impurity phase was generated and the oxygen storage characteristics were degraded. Increase in lattice volume by appropriate amount of La doped led to increase in oxygen storage capacity (Fig. 6).
1. T. Motohashi, Y. Hirano, Y. Masubuchi, K. Oshima, T. Setoyama, S. Kikkawa, Chem. Mater, 2010, vol. 22, 3192.
2. A.J. Wright, H.M. Palmer, P.A. Anderson, G. Greaves, Chem. Mater, 2002, vol. 12, pp. 978-982.
3. A. Ling, R. Zhang, H. Jia, ACSAppl. Mater. Interface, 2015, vol. 7, pp. 14518-14527.
4. G. Xanthopoulou, Adv. Sci. Technol, 2011, vol. 63, pp. 287-96.
5. A. Mukasyan, P. Dinka, J. Self-Propag. High-Temp. Synth., 2007, vol. 16, pp. 23-35.
6. S T. Aruna, A. Mukasyan, Curr. Opin. Solid State Mater. Sci., 2008, vol. 12, pp. 44-50.
7. A. Zhu, A. Nobuta, I. Nakatsugawa, T. Akiyama, Int. J. Hydrogen Energy, 2013, vol. 38, pp.13238-13248.
8. T. Nomura, C. Zhu, N. Sheng, R. Murai, T. Akiyama, J. Alloys Compds, 2015, vol. 645, pp.900-905.
