A determination of the oxygen non-stoichiometry of the oxygen storage material YBaMn2O5+δ

Journal of Solid State Chemistry 230 (2015) 397–403

Contents lists available at ScienceDirect

Journal of Solid State Chemistry journal homepage: www.elsevier.com/locate/jssc

A determination of the oxygen non-stoichiometry of the oxygen storage material YBaMn2O5 þ δ Kannika Jeamjumnunja, Wenquan Gong, Tatyana Makarenko, Allan J. Jacobson n Texas Center for Superconductivity and Department of Chemistry, University of Houston, Houston, TX 77204-5003, USA

art ic l e i nf o

a b s t r a c t

Article history: Received 19 June 2015 Accepted 27 July 2015 Available online 29 July 2015

The A-site ordered double-perovskite oxide, YBaMn2O5 þ δ, has been of recent interest for possible application as an oxygen storage material. In the present study, the oxygen non-stoichiometry of YBaMn2O5 þ δ has been determined as a function of pO2 at 650, 700 and 750 °C by Coulometric titration at near-equilibrium conditions. The results confirm that this perovskite oxide has three distinct phases on oxidation/reduction with δ E0, 0.5 and 1. The stabilities of the YBaMn2O5 þ δ phases span a wide range of oxygen partial pressures (∼10  20 rpO2(atm) r ∼1 ) depending on temperature. The phases interconvert at higher pO2 values at higher temperatures. The partial molar free energies (ΔμO) corresponding to the oxidation of YBaMn2O5 to YBaMn2O5.5 and of YBaMn2O5.5 to YBaMn2O∼6 were determined. The value of ΔμO in both oxidation steps becomes less negative with increasing temperature. At some T and pO2 conditions, YBaMn2O5 þ δ is unstable with respect to decomposition to BaMnO3  δ and YMnO3. This instability is anticipated from the previous studies of the synthesis of YBaMn2O5 þ δ but is more apparent in the present experiments which are necessarily slow in order to achieve equilibrium with respect to the oxygen content. & 2015 Elsevier Inc. All rights reserved.

Keywords: Oxygen storage material Double perovskite oxide Yttrium barium manganese oxide Oxygen non-stoichiometry

1. Introduction Non-stoichiometric oxides which are capable of rapidly and reversibly storing and releasing large amounts of oxygen gas are often called oxygen storage materials (OSMs) or oxygen carriers. These materials are of interest for applications including oxygen enrichment, oxygen separation, syngas production, and catalytic oxidation of hydrocarbons and small molecules [1–3]. Among these applications, chemical looping has recently been suggested as a new technology for clean energy production from coal, partial oxidation of methane, and air separation [4–6]. The availability of suitable oxygen storage materials is a crucial factor for application of this technology. An ideal oxygen storage material should have the following properties: (1) a large oxygen storage capacity (OSC) stored as mobile oxygen ions in the crystal lattice; (2) the uptake/release of oxygen gas should be fast and reversible over a narrow temperature range; (3) adequate phase stability under operating conditions; (4) low operating temperature [4,7]. The best known and most widely used OSMs are the Ce1  xZrxO2  δ compositions with the fluorite structure which have oxygen storage capacities as high as 1500 mmol-O/g [8]. They have been the materials of choice for n

Corresponding author. Fax: þ 1 7137432787. E-mail address: [email protected] (A.J. Jacobson).

http://dx.doi.org/10.1016/j.jssc.2015.07.044 0022-4596/& 2015 Elsevier Inc. All rights reserved.

three way catalysts, air separation, and selective oxidation of methane and have been investigated for solar thermochemical water splitting [1,9–16]. The properties of a number of other oxides have been investigated recently as candidate materials for oxygen storage with reference to the criteria outlined above. Some examples are given in Table 1. Of these, YBaCo4O7 þ δ has a markedly higher oxygen storage capacity than that of Ce1  xZrxO2  δ, but decomposes on heating in an O2 atmosphere above 600 °C. Its phase stability can be improved by partial substitution of Co with Al and Ga, though with a decrease in the oxygen storage capacity [7]. The oxide YBaMn2O5 þ δ with the double perovskite structure has the next highest oxygen storage capacity and reversible oxygen uptake/release characteristics and fast kinetics below 500 °C. The theoretical oxygen storage capacity of the YBaMn2O5–YBaMn2O6 system is  2400 mmol-O/g or  3.85 wt%. The stability of YBaMn2O5 þ δ has been demonstrated by cycling the compound between O2 and 5% H2/95% Ar at 500 °C for over 100 cycles with no detectable decrease in efficiency. YBaMn2O5 þ δ was tested also as a catalyst for CH4 oxidation. Oxidation of CH4 began at around 450 °C and was complete at 550  600 °C [21]. Based on the previous work, the possibility of using YBaMn2O5 þ δ in a chemical looping application gave us an interest in further studying this system. The oxygen uptake/release kinetics of YBaMn2O5 þ δ have been studied by Motohashi et al., who also determined the pO2 dependence of the inter-conversion of

398

K. Jeamjumnunja et al. / Journal of Solid State Chemistry 230 (2015) 397–403

Table 1 Structure type, oxygen storage capacity (OSC) and oxygen uptake/release conditions of some oxygen storage materials. Compound

Type of structure

OSC (lmol-O/ g)

Temperature and atmosphere for oxygen uptake/release processes

Ce1  x ZrxO2  δ [8] YBaCo4O7 þ δ [17] Dy1  xYxMnO3 þ δ [18] Ca2AlMnO5 þ δ [19] LuFe2O4 þ δ [20]

Fluorite Hexagonal Hexagonal

 1500 2700 1200

Oxidation in oxygen atmosphere and reversible reduction in 20% H2 at 500 °C Oxidation at 200  400 °C and reduction at 400 425 °C both in O2 Oxidation under high pressure oxygen at 500 °C and reduction in H2 at 400 °C

Brownmillerite Layered structure of alternating [LuO2] and [Fe2O4] layers Double Perovskite

 1900  1400

Reversible oxygen intake/release at 500  700 °C in O2 Oxidation at 200-500 °C under an oxygen pressure of  0.2x10  3 atm and reduction in H2 from 500 °C

 2400

Oxidation at 200  390 °C in O2 and reduction at 200 490 °C under 5% H2/95% Ar atmosphere

YBaMn2O5 þ δ [21]

YBaMn2O5.5 and YBaMn2O  6 by thermogravimetric analysis in the pO2 range 10–105 Pa (10  4 to 1 atm) and also the range of composition of ‘O6’ [22–24]. We note also that systems with Y partially or completely substituted with other rare-earth elements (La, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, and Er) have been investigated [25–44]. In the present work, the non-stoichiometry of YBaMn2O5 þ δ has been determined by Coulometric titration in sealed electrochemical cells under near equilibrium conditions. Oxygen nonstoichiometry (δ) was measured over a wide range of oxygen partial pressures (∼10  20 rpO2(atm)r ∼1) at 650, 700, and 750 °C, providing both thermodynamic and kinetic data.

2. Experimental section 2.1. Synthesis As previously reported, the synthesis of the double perovskite oxide YBaMn2O5 þ δ must be carried out in a low oxygen partial pressure to prevent the formation of BaMnO3  δ [45–49]. Polycrystalline YBaMn2O5 þ δ was synthesized via a solid state reaction under a mixture of dry N2 and a water-saturated 5% H2/95% Ar gas mixture (pO2 E10  12 atm). As a precursor, a stoichiometric mixture of BaCO3 (Aldrich 99.999%), Y2O3 (Aldrich 99.99%) and Mn2O3 (Aldrich 99%) was ball-milled in isopropanol for 24 h and then dried in an oven for 48 h. About 200 mg of the ball-milled precursor powder was sintered in a tube furnace at 1050 °C for 22 h in the flowing gas mixture described above, and then rapidly cooled to room temperature. The YBaMn2O5 product was annealed at 800 °C for 12 h in flowing O2 gas with 2 °C/min heating and cooling rates to obtain fully-oxidized YBaMn2O5 þ δ. The half-reduced sample (‘O5.5’) was prepared by annealing the fully-oxidized powder under N2 atmosphere at 700 °C for 12 h with 1°C/min heating rate, followed by fast cooling to room temperature (˃20 °C/ min). 2.2. The oxygen non-stoichiometry as a function of T and pO2 The oxygen non-stoichiometry of YBaMn2O5 þ δ was determined by Coulometric titration as a function of pO2 at three temperatures 650, 700, and 750 °C. The experimental method we used has been reported previously and is only briefly reviewed here [50,51]. The Coulometric titration cell is composed of an electrolyte (8-mol% polycrystalline yttria-stabilized zirconia (YSZ) disk), two alumina rings, three glass rings and alumina container. The YSZ disk with Pt mesh electrodes was connected to Pt wires and placed on the top of the titration cell. This YSZ disk was used for pumping oxygen in/out of the cell and also as an oxygen sensor. A Keithley 2400 SourceMeter was used to provide the pump current and to monitor the EMF of the sensor. Air was used as the reference gas. The Pt wires connected to the YSZ disk were

brought out of the cell via glass rings. The gas-tight seals were made by heating the cell above the softening temperature of the glass rings (  750 °C). The height of cell was  15 mm after sealing. An R-type thermocouple was used for monitoring temperature and the experiments were controlled by using LabView software. The oxygen partial pressure (pO2) inside the cell is given by the Nernst equation E ¼–RT/4F ln(pO2cell/pO2ref) where pO2cell and pO2ref are the oxygen partial pressure inside the cell and of the reference gas, respectively, F is the Faraday constant, E is the cell EMF, and R is the gas constant. The non-stoichiometry change (Δδ) was determined in the range ∼10  20 r pO2(atm)r∼1 at 650, 700 and 750 °C. A change in the sensor voltage of less than 0.001% per minute was chosen as the criterion for equilibrium. All titration experiments at each temperature were performed on both decreasing and increasing pO2. The oxygen flux into and out of the cell is controlled by applying a current for a given time [52]. Thus the variation of the number of moles of oxygen transferred by pumping is given by Faraday’s law: ∆nO2, total = it /4F , where i is the applied current, t is the pumping time, and F is the Faraday constant. The change in the total number of moles of oxygen is the sum of the changes in number of moles in the sample and the free space: ∆nO2, total=∆nO2, sample +∆nO2, space . The value of ∆nO2, space was calculated assuming ideal gas behavior. The cell volume is made as small as possible to minimize the correction. The change of oxygen stoichiometry (Δδ) of the sample is expressed by the following equation: ∆δ = 2M /W ∙∆nO2, sample where, M and W are the molecular weight and the weight of the sample. The change of oxygen stoichiometry (Δδ) of the sample is then given by

2M 2M ·∆nO 2, sample = · ∆nO 2, total − ∆nO 2, space W W 2M ⎧ ⎛ it ⎞ ⎛ ∆PV ⎞ ⎫ ⎨⎜ ⎟ − ⎜ ⎟⎬ = W ⎩ ⎝ 4F ⎠ ⎝ RT ⎠ ⎭

∆δ =

(

)

In addition, a small correction is made for the oxygen leakage flux (JO ) through 8-mol% YSZ due to a small electronic contribution to the conductivity at low pO2 using previous results [53]. 2.3. Characterization 2.3.1. Phase purity The phase purity of the samples was verified using powder X-ray diffraction (XRD) with a Phillips PANalytical X’Pert PRO diffractometer with Cu Kα radiation (λ ¼1.54046 Å) at room temperature. The lattice parameters were determined through Le Bail refinement using the GSAS/EXPGUI program. 2.3.2. Chemical analysis The precise oxygen content of the samples was obtained by iodometric titration [47]. Approximately 30 mg of the finely ground sample powder was dissolved in 20 ml of 20% KI solution

K. Jeamjumnunja et al. / Journal of Solid State Chemistry 230 (2015) 397–403

399

and 2 ml of concentrated HCl under a N2 atmosphere. The amount of I2 formed in the reduction reaction of Mn3 þ and/or Mn4 þ to Mn2 þ ion was titrated with 0.02 M Na2S2O3 solution in the presence of a starch indicator. Without the sample, the same amount of KI solution and concentrated HCl was used in a control experiment. Each experiment was repeated for 3 times with good reproducibility. 2.3.3. Thermogravimetric analysis The oxygen uptake/release characteristics of the products were investigated on TA Instruments Hi-Res 2950 thermogravimetric analyzer. For the oxygen uptake study, approximately 40 mg of the as-synthesized YBaMn2O5 þ δ powder was heated and then cooled in a platinum sample pan between room temperature and 800 °C at 2 °C/min in flowing O2 gas. Subsequently, in order to investigate oxygen release process, the oxidized YBaMn2O5 þ δ sample was heated and cooled under a flowing N2 atmosphere in the same temperature range with a rate of 2 °C/min and 20 °C/min, respectively.

3. Results and discussion 3.1. Characterization The control of the oxygen partial pressure is known to be a crucial factor for the synthesis of YBaMn2O5 þ δ with δ ¼0. The compound is formed only under reducing conditions. Polycrystalline YBaMn2O5 was synthesized by solid state reaction at pO2 E10  12 atm and 1050 °C. The product was annealed in O2 and then in N2 to obtain fully-oxidized YBaMn2O6 and half-reduced YBaMn2O5.5 phases, respectively as described previously [21,24,47,54]. The crystal structure of YBaMn2O5 þ δ can be described as an A-site ordered double-perovskite with alternating layers of yttrium and barium ions at the perovskite A-site as previous reported [24,47,49]. The three phases (O5, O5.5 and O  6) of YBaMn2O5 þ δ differ in the occupancy of the oxygen atom site within the yttrium atom layer. The lattice parameters for all of the samples are shown in Table 2. They were single phase, except for one very weak diffraction peak unindexed at  29.5° attributed to the trace of Y2O3 impurity. The diffraction peaks of YBaMn2O5 were indexed on the basis of the tetragonal space group, P4/nmm, giving lattice parameters in good agreement with those in the literature [47]. According to precise neutron diffraction studies reported by Williams et al. [55], the oxidized YBaMn2O6 has a triclinic P-1 space group. In the present work, the X-ray data were refined in the monoclinic space group P2 with the lattice parameters consistent with those previously reported [47]. Nothing in the X-ray data justified lowering the symmetry further. For the N2-annealed product, the diffraction pattern of YBaMn2O5.5 was refined based on an orthorhombic Icma structure, with lattice parameters close to those reported by Perca et al. [49] The refined data for YBaMn2O5.5 are shown in Fig. 1 as a representative example. For comparison with previous work, the oxygen uptake/release characteristics of the present YBaMn2O5 þ δ compounds were

Fig. 1. X-ray data for YBaMn2O5.5. The measured data are shown in red, the calculated data are shown in green, and the difference is shown in pink. Bragg reflections are shown by vertical tick marks.

determined by thermogravimetry. The oxygen content of the assynthesized YBaMn2O5 þ δ determined by iodometric titration corresponded to δ ¼  0.002(1). A sample of YBaMn2O4.998 was heated and then cooled in O2 flowing between 25 and 800 °C at a rate of 2 °C/min. The TGA data (Fig. 2(a)) show that a weight gain occurs in a single sharp step between ∼225 °C and ∼375 °C. The weight gain of 3.59 wt% corresponds to δ ¼0.93 (i.e. YBaMn2O5.93). This magnitude of the weight gain is close to the value (3.85 wt%) expected to obtain a fully oxidized YBaMn2O5 þ δ (with δ ¼ 1). The product from a parallel experiment was shown to be the ‘O6’ phase by X-ray powder diffraction. Subsequently, after cooling to ambient temperature, YBaMn2O5.93 was heated to 800 °C under N2 (pO2 E10  4 atm) at 2 °C/min heating rate, followed by fast cooling (20 °C/min) to room temperature. From the TGA data (Fig. 2(b)), the oxygen release occurs between 400 °C and 800 °C in a single step. The 1.79% weight loss, corresponding to δ ¼0.48 (i.e. YBaMn2O5.45). The XRD pattern of the reduced phase contains only the diffraction lines which are characteristic of the O5.5 phase. Reduction of YBaMn2O5 þ δ under N2 (pO2 E 10  4 atm) involves only O6 and O5.5 phases while only the O5 and O6 phases are observed on oxidation. In general, these results are similar to those reported previously [21,47]. 3.2. Coulometric titration The oxygen non-stoichiometry (δ) of YBaMn2O5 þ δ was measured as a function of oxygen partial pressure (pO2) using Coulometric titration. Measurements were made in sealed electrochemical cells at ∼10  20 rpO2(atm)2 r∼1 on both decreasing and increasing pO2. The dependences of the oxygen non-stoichiometry of YBaMn2O5 þ δ on pO2 at 650, 700 and 750 °C are shown in Fig. 3 (a  c), respectively. As expected, three distinct phases are apparent in the variation of the oxygen content with pO2. At 650 °C, two phases have narrow ranges of composition with oxygen contents (5 þ δ)¼ 4.99-5.02 and 5.46-5.54. The most oxidized phase has a significant range of composition from 5.84  5.95 at 650 °C as shown in Fig. 3(a). Similar behavior in YBaMn2O5 þ δ also occurs at

Table 2 Structural parameters of reduced and oxidized YBaMn2O5 þ δ oxides. Composition

Space group

a (Å )

YBaMn2O5 YBaMn2O5.5 YBaMn2O6

P4/nmm Icma P2

5.5494(1) 8.1590(1) 5.5247(1)

b (Å )

7.5452(1) 5.5192(1) β ¼90.299(0)

c (Å )

Volume (Å 3)

χ2

wRp (%)

7.6522(1) 15.2731(2) 7.6085(1)

235.66(1) 940.29(2) 232.00(1)

2.12 3.65 1.48

6.14 6.66 3.74

400

K. Jeamjumnunja et al. / Journal of Solid State Chemistry 230 (2015) 397–403

104 100

99.5 102

Weight (%)

Weight (%)

103

Δδ = 0.93

101

Δδ = 0.48

99

98.5 100 98 0

0

100 200 300 400 500 600 700 800 o

100 200 300 400 500 600 700 800 o

Temperature ( C)

Temperature ( C)

Fig. 2. Thermogravimetric analysis of YBaMn2O5 þ δ; (a) The ‘O5’ phase was heated from room temperature up to 800 °C in O2 flow; (b) the resulting oxidized product was subsequently heated under N2 up to 800 °C.

higher temperatures, 700 and 750 °C, as presented in Fig. 3(b and c), but the transitions between the phases shift to higher pO2 at higher temperature. The composition ranges of each phase and the pO2 of phase inter-conversion are summarized in Table 3. Note that on reoxidation, steps in the data occur at –log pO2 ¼7.35, 7.71, and 7.77 at 650, 700, and 750 °C (Fig. 3(a  c)), respectively due to some decomposition (see below for further discussion). Consequently some uncertainty exists concerning the upper value of the oxygen content because the experiments were carried out in the temperature sequence 650-700-750 °C and some decomposition occurred during reoxidation from 650 °C before the reduction experiment at 700 °C and similarly during reoxidation at 700 °C

before the reduction experiment at 750 °C. This has a small effect on the upper value of the oxygen content because the average Mn oxidation state for YBaMn2O6 and the phase separated products is the same at high pO2 (  1 atm). The value of pO2 at which 5.5 þ δ converts to 5.5 is not effected and the values scale linearly with temperature (Fig. 4). The pO2 dependence on temperature of O5 to O5.5 and O5.5 to O6 phase transition is shown in Fig. 4. The present results for the O5.5 to O6 inter-conversion can be compared with the recent TGA measurements of Motohashi et al. [22] Based on their results, the pO2 of the transition between O5.5 and O6 phases occurred approximately at –log pO2 ¼3.40, 2.71, 2.46 at 650, 700 and 750 °C,

6.0

6.0

reducing oxidizing

oxidizing

δ

5+δ δ in YBaMn O δ 2 5+δ

5.8

2

5+δ δ in YBaMn O

5+δ δ

reducing

5.6 5.4

-log(pO ) = 7.35 2

5.2 5.0

5.8 5.6 5.4

-log(pO ) = 7.71 2

5.2 5.0

20 18 15 13 10

8

5

3

0

20 18 15 13 10

-log (pO ), pO in (atm) 2

8

5

3

0

-log (pO ), pO in (atm)

2

2

2

5.8

reducing oxidizing

2

5+δ δ in YBaMn O

5+δ δ

6.0

5.6 5.4

-log(pO ) = 7.77 2

5.2 5.0 20 18 15 13 10

8

5

3

0

-log (pO ), pO in (atm) 2

2

Fig. 3. The pO2 dependence of oxygen-non stoichiometry on pO2 measured on reduction and reoxidation of YBaMn2O5 þ δ at (a) 650 °C, (b) 700 °C and (c) 750 °C.

K. Jeamjumnunja et al. / Journal of Solid State Chemistry 230 (2015) 397–403

401

Table 3 Composition ranges of each phase and the pO2 of phase inter-conversion in YBaMn2O5 þ δ at 650, 700 and 750 °C. Temperature (°C) O5 phase composition range

-log pO2 of the transition between O5 and O5.5

O5.5 phase composition range

-log pO2 of the transition between O5.5 and O6

O6 phase composition rangea

650 700 750

14.82 13.41 12.18

5.46–5.54 5.47–5.55 5.43–5.54

3.23 2.83 2.51

5.84–5.95 5.83–5.95 5.75–5.87

a

4.99–5.02 5.00–5.07 4.99–5.11

Some uncertainty exists concerning the upper value of the oxygen content (see text)

-2 Transition of O to O

-2.5

Transition of O

5.5

5.5

to O

6

-3

2

log (pO ), pO in atm

5

2

-3.5 -12 -13 -14 -15 -16 600

650

700

750

800

o

Temperature ( C) Fig. 4. The temperature dependence of the pO2 corresponding to the transitions from O5 to O5.5 and O5.5 to O5.5 þ δ. The data were measured on reduction.

respectively, in good agreement with the present data given in Table 3.

are the steps in that data that occur at –log pO2 ¼7.35, 7.71, and 7.77 at 650, 700, and 750 °C (Fig. 3(a c)), respectively. The kinetics associated with these discontinuities are extremely slow. Eventually the data rejoin the curve obtained on reduction at 650 and 700 °C but at 750 °C it was necessary to relax the criterion for equilibrium to obtain a fully oxidized sample. The system is not completely reversible on re-oxidation due to phase separation as evidenced by the slow kinetics. The oxide phases after the experiments at 700 and 750 °C were removed from the cell and analyzed by powder X-ray diffraction. The diffraction patterns confirmed that partial and complete decomposition to BaMnO3  δ and YMnO3 occurred at 700 and 750 °C, respectively, as shown in Fig. 6 for the product obtained at 750 °C. At higher temperature (800 °C) it was not possible to reach equilibrium in the reduction reaction starting at O6 because phase separation to BaMnO3  δ and YMnO3 occurred at the high pO2 and 800 °C. Based on literature data [56], we expect δ E 0 for BaMnO3  δ and consequently the average Mn oxidation state and total oxygen content is the same for YBaMn2O6 and for the phase separated products at high pO2 (  1 atm). 3.4. Thermodynamics

3.3. Kinetic effects and phase stability With the chosen criterion for equilibrium (0.001% change in the sensor voltage per minute) different kinetic regimes were observed. As an example, the equilibrium times for oxidation and reduction are shown in Fig. 5 as a function of oxygen non-stoichiometry at 650 °C. On reduction, equilibrium times were ∼5 h per point except in the vicinity of the inter-conversion of O5 and O5.5 where they were significantly longer (∼15 h per point). This is consistent with the observation that O5.5 is not observed on rapid oxidation of O5 in oxygen (slow inter-conversion of O5 and O5.5). An unusual feature

The partial molar free energy of oxygen atom (ΔμO) at equilibrium between the solid and the gas phase calculated from the reduction data in Fig. 3 is shown in Fig. 7. The data clearly show the existence of two two-phase regions corresponding to the reactions: 2YBaMn2O5.0 (s) þ1/2O2 (g)-2YBaMn2O5.5 (s)

(1)

10/3YBaMn2O5.5 (s)þ 1/2O2 (g)-10/3YBaMn2O5.8 (s)

(2)

YBaMn O 2

o

Equilibrium time (hour)

650 C

reducing oxidizing

80 60

-log(pO ) = 7.35 2

40

Intensity (a.u.)

100

5+δ δ

2H-BaMnO

3

YMnO

3

20 0 6.0

10 20 30 40 50 60 70 80 90 5.8

5.6

5.4

5.2

5.0

5+δ δ in YBaMn O 2 5+δ δ

δ

Fig. 5. Equilibrium times as a function of oxygen non-stoichiometry for YBaMn2O5 þ δ at 650 °C.

2θ (deg) Fig. 6. X-ray data for YBaMn2O5 þ δ collected after re-oxidation by Coulometric titration at 750 °C, compared with the X-ray diffraction patterns of 2H-BaMnO3 and YMnO3 obtained from the ICSD database.

402

K. Jeamjumnunja et al. / Journal of Solid State Chemistry 230 (2015) 397–403

Δμ O (kJ/mol)

0

-50

-100

-150

-200 6.0

5.8

5.6

5.4

5.2

5+δ δ in YBaMn O

0

0

-50

-50

Δμ (kJ/mol) O

Δμ O (kJ/mol)

2

-100

5+δ δ

-100

-150

-150

-200 6.0

5.0

5.8

5.6

5.4

5.2

-200 6.0

5.0

5+δ δ in YBaMn O 2

5.8

5.6

5.4

5.2

5+δ δ in YBaMn O

5+δ δ

2

5.0

5+δ δ

Fig. 7. Partial molar free energy of oxygen atom (ΔμO) as a function of oxygen content for YBaMn2O5 þ δ at (a) 650 °C, (b) 700 °C and (c) 750 °C.

and the existence of a single phase with a range of stoichiometry above YBaMn2O5.8. At 700 and 750 °C, YBaMn2O5.0 also appears to have a small range of stoichiometry (  5.0 to 5.1 see Table 3). At 650 °C, the average values of ΔμO in the two phase regions equal  131.00 and 28.60 kJ/mol for oxidation from δ ¼ 0 (YBaMn2O5) to δ ¼0.5 (YBaMn2O5.5), and from δ ¼0.5 (YBaMn2O5.5) to δ E0.8 (YBaMn2O5.8), respectively, as shown in Fig. 7(a). Both reactions are exergonic. In addition, the partial molar free energy on oxidation is less negative with increasing temperature as shown in Fig. 7(b) and (c). The value of ΔμO in oxidation from δ ¼ 0 (YBaMn2O5) to δ ¼0.5 (YBaMn2O5.5) is ΔμO ¼  124.97 kJ/mol at 700 °C and ΔμO ¼  119.34 kJ/mol at 750 °C. For oxygen intake from δ ¼0.5 (YBaMn2O5.5) to δ E0.8 (YBaMn2O5.8), ΔμO becomes less negative from –26.39 kJ/mol at 700°C to –24.64 kJ/mol at 750 °C. Plots of R/2  ln(pO2) versus 1/T and RT/2  ln(pO2) versus T are linear indicating that the enthalpy and entropy are constant in this narrow temperature range. The values for the changes in enthalpy and entropy are  238.7 kJmol  1 and  116.6 J mol  1 K  1 for the conversion of O5 to O5.5 (reaction 1), and  39.2 kJ mol  1 and 23.8 J mol  1 K  1 for the conversion of O5.5 to O5.8 (reaction 2).

oxygen non-stoichiometry of YBaMn2O5 þ δ was determined as a function of pO2 and temperature by using a Coulometric titration method. Only three distinct phases of YBaMn2O5 þ δ with δ E 0, 0.5 and 1 are observed during the oxidation/reduction process. The phase transitions shift to higher pO2 as the temperature increases. The partial molar free energies (ΔμO) corresponding to the oxidation of YBaMn2O5 to YBaMn2O5.5 and of YBaMn2O5.5 to YBaMn2O5.8 were determined. The value of ΔμO in both oxidation steps becomes less negative with increasing temperature. The YBaMn2O5 þ δ compound has possible applications in oxygen storage technology but based on the present results, the instability with respect to decomposition to BaMnO3  δ and YMnO3 at high pO2 and temperature may be of some concern for practical applications.

Acknowledgments The work was supported by the Robert A. Welch Foundation (Grant no. E-0024, TM; TGA measurements) and the U.S. Department of Energy (U.S. DOE), Office of Basic Energy Sciences Division of Materials Sciences and Engineering (under Award no. DESC0001284, KJ, WG, AJJ; experiments and analysis).

4. Conclusions References The A site-ordered double-perovskite YBaMn2O5 þ δ was successfully synthesized by solid state reaction under a reducing atmosphere. Thermogravimetric measurements show good oxygen uptake/release ability of this compound at moderate temperatures in agreement with previous work. The oxygen storage capacity is ∼2200 μmol-O/g and is higher than that of Ce1  xZrxO2 þ δ. The

[1] [2] [3] [4] [5] [6]

J. Kašpar, P. Fornasiero, J. Solid State Chem. 171 (2003) 19–29. Z. Yang, Y.S. Lin, Y. Zeng, Ind. Eng. Chem. Res. 41 (2002) 2775–2784. Y. Wei, H. Wang, F. He, X. Ao, C. Zhang, J. Nat. Gas Chem. 16 (2007) 6–11. K. Li, H. Wang, Y. Wei, J. Chem. (2013), Article ID 294817. K. Wang, Q. Yu, Q. Qin, J. Therm. Anal. Calorim. 112 (2013) 747–753. S. Bhavsas, G. Veser, RSC Adv. 4 (2014) 47254–47267.

K. Jeamjumnunja et al. / Journal of Solid State Chemistry 230 (2015) 397–403

[7] O. Parkkima, H. Yamauchi, M. Karppinen, Chem. Mater. 25 (4) (2013) 599–604. [8] Y. Nagai, T. Yamamoto, T. Tanaka, S. Yoshida, T. Nonaka, T. Okamoto, A. Suda, M. Sugiura, Catal. Today 74 (2002) 225–234. [9] S. Damyanova, B. Pawelec, K. Arishtirova, M.V. Martinez Huerta, J.L.G. Fierro, Appl. Catal. A Gen. 337 (2008) 86–96. [10] R. DiMonte, P. Fornasiero, M. Graziani, J. Kašpar, J. Alloys Compd. 275  277 (1998) 877–885. [11] H. He, H.X. Dai, C.T. Au, Catal. Today 90 (2004) 245–254. [12] J. Kašpar, P. Fornasiero, N. Hickey, Catal. Today 77 (2003) 419–449. [13] K. Otsuka, Y. Wang, M. Nakamura, Appl. Catal. A Gen. 183 (1999) 317–324. [14] P. Pantu, K. Kim, G.R. Gavalas, Appl. Catal. A Gen. 193 (2000) 203–214. [15] M. Sugiura, Catal. Surv. Asia 7 (1) (2003) 77–87. [16] Y. Hao, C.-K. Yang, S.M. Haile, Chem. Mater. 26 (2014) 6073–6083 , and references therein. [17] M. Karppinen, H. Yamauchi, S. Otani, T. Fujita, T. Motohashi, Y.-H. Huang, M. Valkeapää, H. Fjellvåg, Chem. Mater. 18 (2006) 490–494. [18] S. Remsen, B. Dabrowski, Chem. Mater. 23 (2011) 3818–3827. [19] T. Motohashi, Y. Hirano, Y. Masubuchi, K. Oshima, T. Setoyama, S. Kikkawa, Chem. Mater. 25 (2013) 372–377. [20] M. Hervieu, A. Guesdon, J. Bourgeois, E. Elkaїm, M. Poienar, F. Damay, J. Rouquette, A. Maignan, C. Martin, Nat. Mater. 13 (2013) 74–80. [21] T. Motohashi, T. Ueda, Y. Masubuchi, M. Takiguchi, T. Setoyama, K. Oshima, S. Kikkawa, Chem. Mater. 22 (2010) 3192–3196. [22] T. Motohashi, T. Takahashi, M. Kimura, Y. Masubuchi, S. Kikkawa, Y. Kubota, Y. Kobayashi, H. Kageyama, M. Takata, S. Kitagawa, R. Matsuda, J. Phys. Chem. C 119 (2015) 2356–2363. [23] T. Motohashi, T. Ueda, Y. Masubuchi, S. Kikkawa, J. Ceram. Soc. Jpn. 119 (11) (2011) 894–897. [24] T. Motohashi, T. Ueda, Y. Masubuchi, S. Kikkawa, J. Phys. Chem. C 117 (2013) 12560–12566. [25] D. Ávila-Brande, G. King, E. Urones-Garrote, Subakti, A. Llobet, S. GarcíaMartín, Adv. Funct. Mater. 24 (2014) 2510–2517. [26] V. Caignaert, F. Millange, B. Domengès, B. Raveau, Chem. Mater. 11 (1999) 930–938. [27] A. Klimkowicz, K. Świerczek, A. Takasaki, J. Molenda, B. Dabrowski, Mater. Res. Bull. 65 (2015) 116–122. [28] A. Klimkowicz, K. Świerczek, K. Zheng, M. Baranowska, A. Takasaki, B. Dabrowski, Solid State Ionics 262 (2014) 659–663. [29] A. Klimkowicz, K. Zheng, G. Fiolka, K. Świerczek, CHEMIK 67 (12) (2013) 1199–1206. [30] F. Millange, V. Caignaert, B. Domengès, B. Raveau, Chem. Mater. 10 (1998)

403

1974–1983. [31] T. Motohashi, M. Kimura, T. Inayoshi, T. Ueda, Y. Masubuchi, S. Kikkawa, Dalton Trans. (2015), http://dx.doi.org/10.1039/C4DT03863K, Advance Article. [32] T. Nakajima, H. Kageyama, H. Yoshizawa, K. Ohoyama, Y. Ueda, J. Phys. Soc. Jpn. 72 (12) (2003) 3237–3242. [33] T. Nakajima, H. Kageyama, H. Yoshizawa, Y. Ueda, J. Phys. Soc. Jpn. 71 (12) (2002) 2843–2846. [34] T. Nakajima, H. Yoshizawa, Y. Ueda, J. Phys. Soc. Jpn. 73 (8) (2004) 2283–2291. [35] S. Sengodan, S. Choi, A. Jun, T.H. Shin, Y.-W. Ju, H.Y. Jeong, j Shin, J.T.S. Irvine, G. Kim, Nat. Mater. 14 (2) (2015) 205–209. [36] K. Świerczek, A. Klimkowicz, A. Niemczyk, A. Olszewska, T. Rząsa, J. Molenda, A. Takasaki, Funct. Mater. Lett. 7 (6) (2014) 1440004. [37] K. Świerczek, A. Klimkowicz, K. Zheng, B. Dabrowski, J. Solid State Chem. 203 (2013) 68–73. [38] A.A. Taksin, Y. Ando, Phys. Rev. Lett. 98 (2007) 207201. [39] A.A. Taksin, A.N. Lavrov, Y. Ando, Appl. Phys. Lett. 86 (2005) 091910. [40] I.O. Troyanchuk, S.V. Trukhanov, G.M. Chobot, H. Szymczak, Phys. Solid State 45 (2) (2003) 276–282. [41] I.O. Troyanchuk, S.V. Trukhanov, G. Szymczak, Crystallogr. Rep. 47 (4) (2002) 658–665. [42] Y. Ueda, T. Nakajima, J. Phys.: Condens. Matter 16 (2004) S573–S583. [43] A.J. Williams, J.P. Attfield, Phys. Rev. B. 66 (2002) 220405. [44] A.J. Williams, J.P. Attfield, Phys. Rev. B. 72 (2005) 184426. [45] T.P. Beales, M. Mӧlgg, J. Jutson, C.M. Friend, Phys. Stat. Sol. 161 (a) (1997) 271–282. [46] J.P. Chapman, J.P. Attfield, M. Molgg, C.M. Friend, T.P. Beales, Angew. Chem., Int. Ed. 35 (1996) 2482–2484. [47] M. Karppinen, H. Okamoto, H. Fjellvåg, T. Motohashi, H.J. Yamauchi, Solid State Chem. 177 (2004) 2122–2128. [48] J.A. Mcallister, J.P. Attfield, J. Mater. Chem. 8 (1998) 1291–1294. [49] C. Perca, L. Pinsard-Gaudart, A. Daoud-Aladine, M.T. Fernández-Díaz, Rodríguez-Carvajal, J. Chem. Mater. 17 (7) (2005) 1835–1843. [50] C.Y. Park, A.J. Jacobson, J. Electrochem. Soc. 152 (2005) J65–J73. [51] J. Yoo, C.Y. Park, A. Jacobson, J. Solid State Ionics 175 (2004) 55–58. [52] K. Kosuge, Chemistry of Non-Stoichiometric Compounds, Oxford University Press, Oxford, 1994. [53] J.H. Park, R.N. Blumenthal, J. Electrochem. Soc. 136 (1989) 2867–2876. [54] F. Millange, E. Suard, V. Caignaert, B. Raveau, Mater. Res. Bull. 34 (1) (1999) 1–9. [55] A.J. Williams, J.P. Attfield, Phys. Rev. B 72 (2005) 024436. [56] T. Negas, R.S. Roth, J. Solid State Chem. 3 (1971) 323–339.