Electronic/Ionic Conductivity and Oxygen Diffusion

Coefficient of Sr-Fe-Co-0 System*

B. Ma, J. H. Park, and U. Balachandran Energy Technology Division, Argonne National Laboratory

Argonne, IL 60439 USA

C. U. Segre Department of Physics, Illinois Institute of Technology

Chicago, IL 60616 USA

March 1995

under contract No. W-31-104ENG-38.

nonexclusive, royalty-free license to publish or reproduce the published form of thls contribution, or allow others to do EO, for

Paper to be published in the proceedings of the Material Research Society Spring Meeting, San Francisco, CA, April 17-21, 1995

* Work at Argonne National Laboratory is supported by the U.S. Department of Energy, Pittsburgh Energy Technology Center, under contract W-3 1- 109-Eng-38.

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, make any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

DlSCLAlMER

Portions of this document may be illegible in electronic image products. lrnages are produced from the best available original document.

Electronic/Ionic Conductivity and Oxygen Diffusion

Coefficient of Sr-Fe-Co-0 System*

B. Ma, J. H. Park, and U. Balachandran Argonne National Laboratory, Argonne, IL 60439 USA

C. U. Segre Department of Physics, Illinois Institute of Technology

Chicago, IL 60616 USA

ABSTRACT

Oxides in the system Sr-Fe-Co-0 exhibit both electronic and ionic conductivities. Recently, Sr-Fe-Co-0 system attracted great attention because of the potential to be used for oxygen permeable membranes that can operate without the electrodes or external electrical circuitry. Electronic and ionic conductivities at various temperatures have been measured on two compositions in Sr-Fe-Co-0 system named SFC-1 and SFC-2. The electronic transference number is much greater than the ionic transference number in SFC-1 sample, while the electronic and ionic transference numbers are very close in SFC-2 sample. At 8OO0C, the electronic conductivity and ionic conductivity are -76 S-cm-l and =4 S-cm-1, respectively, for SFC-1. While, for SFC-2, the electronic and ionic conductivities are =lo S-cm-1 and =7 S-cm-1, respectively. By a local fitting to o-T = A exp(-Ea/kT), we found that the oxide ion activation energies are 0.92 eV and 0.37 eV respectively for SFC-1 and SFC-2 samples. Oxygen diffusion coefficient of SFC-2 is = 9 ~ 1 0 - ~ cm2/sec at 9 0 0 0 ~ .

- INTRODUCTION

Mixed conductors find wide application in high temperature solid-state electrochemical devices such as solid oxide fuel cells, batteries and sensors. The same materials also hold particular promise as ceramic membranes designed to separate oxygen from air, being impervious to other gaseous constituents. High oxygen permeability, usually comes along with high oxygen ionic conductivity, is desired in the separation process. (La,Sr)(Fe,Co)Ox systems have been shown by Teraoka et al. [ 1, 21 to have not only mixed (electronic and ionic) conductivities but also appreciable oxygen permeability (two orders of magnitude higher than that of stabilized zirconia at 800°C). One could use this material for producing syngas (CO + H2) by direct conversion of methane and other basic hydrocarbon gas, such as coal gas. Air can be used as oxidant in the conversion process because the dense ceramic membrane made from this material can successfully separate oxygen from air at flux rate that could be considered commercially feasible. Thus these materials can potentially be used in applications for improving the economics for methane conversion processes [3-71.

Electronic and ionic transference numbers are important parameters for understanding the thermodynamic process in ceramic material. Four-probe electron-blocking method 181, used in our experiments, has been shown to give reliable data of oxide ion conductivity for mixed conductors.

I n this paper, we discuss the temperature dependence properties of two compositions in the Sr-Fe-Co-0 systems, named as SFC-1 and SFC- 2. Activation energies, Ea, are obtained by a local fitting to 0 T = A exp(-E, / kT) . A comparison with the electronic/ionic conductivity of other compositions in the (La,Sr)(Fe,Co)Ox systems is also provided in this paper.

EXPERIMENTAL

- Two ceramic powders in the system Sr-Fe-Co-0 with different stoichiometries, designated SFC-1 and SFC-2, were made by solid state reaction of the constituent cation salts. The stoichometry of SFC-1 is SrFe0.2Co0.80~, which has been reported by Teraoka et al. [ 1, 21. SFC-2, an improved version of SFC-1, has stoichometry of SrFeCo0.50,. I t also has better mechanical properties than SFC-1 [9, 101. SFC-1 and SFC-2 powders were made using appropriate amounts of SrC03, Co(N03)2*6H20, and Fe2O3, mixing and grinding in isopropanol with 21-02 media for 15 h. After drying, the mixtures were calcined in air at 850°C for 16 h with intermittent grinding. After final calcination, the powders were ground with an agate mortar and pestle to an average size of =7 pm. X-ray diffraction (XRD), Scanning electron microscopy (SEM), thermal analysis results and particle size distribution analysis were reported earlier [9, 1 11.

The resulting powders were pressed with 1 . 2 ~ 1 0 ~ MPa load into pellets of 21.5 mm in diameter and =3 mm in thickness. Pellets were sintered in air at =12OO"C for 5 h. The pellets were cut into small bars used for conductivities measurement. The bulk densities of samples are -95% of their theoretical values.

The experimental set up for the total conductivity measurement is shown in Fig.1. Pt wires were used as probes in the four-probe measurement. The resistance of specimen was measured with HP 4192A LF impedance analyzer. At low frequency, the measured resistance was the same value as that got with conventional dc method. For the measurement of oxygen ion conduction, yttria stabilized zirconia (YSZ, with 8 mol.% Y2O3) was used for electron (hole) blocking electrode. A schematic drawing of experimental arrangement is shown in Fig. 2. Gold paste was used in between YSZ and sample to eliminate the contact resistance. Experiments were carried out in flowing air environment. At each measuring temperature, we waited three hours for the specimen to reach equilibrium before taking data.

RESULTS AND DISCUSSION

Conductivities were calculated from following equation:

o=- d v v R - S

where dm and S are the separation of voltage probes and the cross- sectional area of the specimen, R the resistance measured with 4-probe method.

The measured total conductivities of SFC-1 and SFC-2 in air at various temperatures are shown in Fig. 3. Total conductivity in SFC-1 decreases with increasing temperature and behaves metallic. However, for SFC-2 the total conductivity increases with increasing temperature, behaves like semiconductor. At 8OO0C, the total conductivities for SFC-1 and SFC-2 are 80 Sscrn-1 and 17 S-cm-1, respectively.

The measured ionic conductivities of SFC-1 and SFC-2 in air at various temperatures are shown in Fig. 4. Different from the total conductivity behavior, the ionic conductivities increase with increase in temperature for both SFC-1 and SFC-2 samples. At 8OO0C, the ionic conductivities for SFC-1 and SFC-2 are 4 and 7 S-cm-1, respectively.

Since the total conductivity (cTtotal), is the summation of ionic conductivity (Gi), and electronic conductivity (eel),

we can obtain the electronic conductivity by using equation (2). At 8OO0C, the electronic conductivities of SFC-1 and SFC-2 samples are 76 and 10 Seem-1 respectively.

Temperature dependence of ionic transference numbers, t i (defined as ti = Oi/cTfotd 1, are shown in Fig. 5. For SFC- 1, ti increases linearly with temperature; for SFC-2, ti is more or less independent of temperature. The ionic transference number of SFC-2 is =lo times greater than that of SFC-1 at 800°C. For SFC-1, the ratio of electronic transference number (ter) to ionic transference number, tel/ti, is about 10 at 800°C.

This indicates that electronic conductivity behavior dominates the total conductivity in SFC- 1 while in SFC-2 electronic and ionic conductivities play almost equivalent roles. This might be the reason why SFC-1 has completely different temperature dependence behavior for electronic and ionic conductivities while SFC-2 has very similar temperature dependence behavior for both. Comparing to other systems like (La,Sr)(Fe,Co)03 [8, 12-14], Sr-Fe-Co-0 system shows quite high ionic conductivity. Table 1 shows the collected information on the electrical conductivities for the (La,Sr)(Fe,Co)Ox systems.

-

Table 1. Conductivities measured in air at 800°C for (La,Sr)(Fe,Co)Ox systems

Samples Electronic Ionic (Ti Method for (Ti Ref* (S-cm-1 ) ( S-cm-1 ) measurement 76 4 4-terminal, YSZ *

electron block SFC- 1

SFC-2 10 7 4-terminal, YSZ ,, electron block

~ . 6 ~ ~ 0 . 4 ~ ~ 0 . 2 ~ ~ 0 . 8 ~ 3 300 0.01 4-terminal, YSZ 13 electron block

%. 6 ~ ~ 0 . 4 ~ ~ 0 . 2 ~ ~ 0 . 8 ~ 3 300 0.003 2 -terminal. 1 4 electron block

%.SsrO.2~0.8Fe0.203 600 15 4-terminal, YSZ 12 electron block

%.8~~0.2~~0.8F~0.203 250 0.10 4-terminal, YSZ 8 electron block

h.75Sr0.25Fd3 50 0.03 1*0/160 exchange 15

~ ~~

* present work.

Motion of oxide ions can be described by a thermally activated process for which the ionic conductivity satisfies [ 161

where A is a constant, Ea is the activation energy for oxide ion, k, Boltzmann constant (8.63 x loq5 eV/K), and T the absolute temperature. Ln(T.oi) us. 10000/T for SFC-1 and SFC-2 are plotted in Fig. 6. By doing a local fitting to equation (3), we found that the oxide ion activation energy for SFC-1 and SFC-2 are 0.92 eV and 0.37 eV, respectively. The activation energy we obtained for SFC-1 is consistent with that reported by Teraoka et a2. [ 171, but smaller than that got by Nisancioglu et al. [ 181 from their oxygen diffusion coefficients data. SFC-2 sample has a oxide ion activation energy which is much lower than the activation energy of common ceramic oxide, which is about 1 eV. It implies that oxide ion can move more easily in SFC-2 sample. From another point of view, oxide ions are more likely to have uniform distribution in SFC-2 sample rather than in SFC-1 sample.

The oxygen diffusion coefficient of SFC-2 has been measured by conductivity relaxation methed [19]. Because of the fracture of SFC-1 at low oxygen partial pressure, we were not be able to perform relaxation experiment on SFC-1 sample. The diffusion coefficient of SFC-2 is ~9x10-7 cm2/sec at 900°C.

Membrane reactor tubes from SFC-1 and SFC-2 ceramic powders have been fabricated via an extrusion technique [3,6,9, lo]. These tubes were tested in actual methane gas conversion reactors. Methane gas was passed on one side and air on other side of the ceramic tubes. Reactor tubes made out of SFC-1 composition disintegrated within minutes into conversion reaction, whereas reactor tubes made out of SFC-2 composition were tested for over 1000 h without any fracture. Methane conversion efficiency > 98% were obtained in reactor runs over 1000 h at 850°C [6, 9, 101.

CONCLUSION

SFC-2 has different electronic ionic properties from SFC-1. The ionic conductivity and electronic conductivity we got for SFC-2 sample are very close in value. However, the electronic conductivity of SFC-1 is

much greater than its ionic conductivity. SFC-2 has more stable structure [6, 101 than SFC-1. Tubes made with SFC-2 provided methane-conversion efficiency of greater than 98%, and have been in operation for more than 1000 h. The activation energy of SFC-2 is smaller than that of SFC-1. Suggesting that the migration of oxide ion in SFC-2 sample is easier than that in SFC-1 sample. Our results indicate that SFC-2 is a better material for making oxygen permeation membranes.

-

ACKNOWLEDGMENTS

Work at Argonne National Laboratory is supported by the U.S. Department of Energy, Pittsburgh Energy Technology Center, under contract W-3 1- 109-Eng-38.

REFERENCES

1. Y. Teraoka, H. M. Zhang, S. Furukawa, and N. Yamozoe, Chem Lett., 1985, 1743, 1985.

2. Y. Teraoka, T. Nobunaga, and N. Yamazoe, Chem. Lett., 1988, 503, 1988.

3. U. Balachandran, S. L. Morissette, J. J. Picciolo, J. T. Dusek, R. B. Poeppel, S. Pei, M. S. Kleefisch, R. L. Mieville, T. P. Kobylinski, and C. A. Udovich, Roc. Int. Gas Research Con$, (H. A. Thompson ed.) pp. 565-573, Government Institutes, Inc., Rockville, MD, 1992.

4. T. J. Mazanec, T. L. Cable, and J. G. Jr. Frye, Solid State lonics, 111, 53, 1992.

5. A. C. Bose, J. G. Stigel, and R. D. Srivastava, "Gas to Liquids Research Program of the U.S. Department of Energy: Programmatic Overview," paper was presented at the Symposium on Alternative Routes for the

Production of Fuels, American Chemical Society National Meeting, Washington DC, August 21-26, 1994.

6. U. Balachandran, S. L. Morissette, J. T. Dusek, R. L. Mieville, R. B. Poeppel, M. S . Kleefisch, S . Pei, T. P. Kobylinski, and C. A. Udovich, Proceedings of Coal liquefaction and Gas Conversion Contractors Review Conference (S . Rogers et al. ed.), Vol. 1, pp. 138-160, U.S. Department of Energy, Pittsburgh Energy Technology Center, Pittsburgh, PA, September 27-29, 1993.

7. T. L. Cable, European Patent EP 0438 902 A2, July 31, 1991.

8. Y. Teraoka, H. M. Zhang, K. Okamoto, and N. Yamazoe, Mat. Res. BuL, 23, 51, 1988.

9. U. Balachandran, J. T. Dusek, S . M. Sweeney, R. L. Mieville, P. S . Maiya, M. S . Kleefisch, S. Pei, T. P. Kobylinski, and A. C. Bose, submitted for presentation at the 3rd International conference on Inorganic Membranes, July 10- 14, Worcester, MA, USA, 1994.

10. U. Balachandran, J. T. Dusek, S . M. Sweeney, R. B. Poeppel, R. L. Mieville, P. S. Maiya, M. S. Kleefisch, S. Pei, T. P. Kobylinski, C. A. Udovich, and A. C. Bose, American Ceramic Society BulZetin, 74, 71, 1995.

11. S. Pei, M. S . Kleefisch, T. P. Kobylinski, J. Faber, C. A. Udovich, V. Zhang-McCoy, B. Dabrowski, U. Balachandran, R. L. Mieville, and R. B. Poeppel, Catalysis Lett., 30, 201-212, 1995.

12. W. L. Worrell, P. Han, and J. Huang, in "High Temperature Electrochemical Behavior of Fast Ion and Mixed Conductors", ( F. W. Poulsen, J. J. Bertzen, T. Jacobson, E. Skou, and M. J. L. Ostergood, ed. ), Riss National Laboratory, pp. 461-466, 1993.

13. H. U. Anderson, C. C. Chen, L. W. Tai, and M. M. Nasrallah, Published in "Proceedings of the 2nd International symposium on Ionic and Mixed Conduction Ceramics", (T.A. Ramanarayaman, W. L. Worrell and H. L. Tuller, ed.), pp. 376-387, The Electrochem. SOC., 1994.

14. C. C. Chen, M. M. Nasrallah, and H. U. Anderson, Submitted to J. Electrochem. SOC., 1994. -

15. T. Ishigaki, S. Yamauchi, K. Kishio, J. Mizusaki, and K. Fuek, Solid State Chem, 73, 179, 1988.

16. See for example, H. L. Tuller in "Nonstoichiometric Oxides" (0. T. Sorensen ed.), pp. 276, Academic Press, New York, 1981.

17. Y. Teraoka, T. Nobunaga, K. Okamoto, N. Miura, and N. Yarnazoe, Solid State lonics, 48, 207, 1991.

18. K. Niscancioglu and T. M. Gur, to be published in Solid State lonics, 1994.

19. B. Ma, U. Balachandran, J. H. Park, and C. U. Segre, to be published in Solid State lonics.

Thermocouple

Brass Cap

---+ Gasoutlet

Quartz Tube _cc_

Gas inlet 1

Gas inlet 2

x Electric Furnace

Specimen

F'igure 1. Schematic drawing of experimental arrangement for high-temperature conductivity measurement.

\

I+

Pt electrod \

Figure 2. Experimental setup of 4-probe electron-block technique.

13

160

140

120

100

80

60

4 0

2 0

0

Figure 3.

0

0 0

0 0

0- 0

0 0

0 0 0

500 600 700 800 900 1000

Temperature ("C)

Temperature dependence of total conductivity. Open circle, represents SFC- 1 ; open square, represents SFC-2.

10

8

n r

€ 0 v) Y .- b

6

0

4

2

0

Figure 4.

650

0 0

0 0

0

0

700

0 0

0

0

0

0

750 800

Temperature ( O C )

850

Temperature dependence of ionic conductivity. Open circle, represents SFC- 1 ; open square, represents SFC-2.

+;-

0.4

0.3

0.2

0.1

0

650 700 750 800 850 900 950

Temperature ( O C )

Figure 5. Temperature dependence of ionic transference number. Open circle, represents SFC- 1 ; open square, represents SFC-2.

1 1

1 0

9.5

9.0

8.5

8.0

7.5

7.0

6.5

8 8.5 9 9.5 1 0

10000/T(K)

10.5 1 1

. ,

Figure 6. Ln(T(a a) vs. reciprocal of temperature. Open circle, represents SFC- 1; open square, represents SFC-2.