The catalytic activity of rare earth manganites

The catalytic activity of rare earth manganites

Mat. Res. Bull., Vol. 15, pp. 269-273, 1980. P r i n t e d in the USA. 0025-5408/80/020269-05502.00/0 Copyright (c) 1980 Pergamon P r e s s Ltd. THE ...

250KB Sizes 5 Downloads 14 Views

Mat. Res. Bull., Vol. 15, pp. 269-273, 1980. P r i n t e d in the USA. 0025-5408/80/020269-05502.00/0 Copyright (c) 1980 Pergamon P r e s s Ltd.




T. Arakawa, A. Yoshida and J. Shiokawa Department of Applied Chemistry, Faculty of Engineering, Osaka University, Yamadakami, Suita-Shi, Osaka-Fu, Japan

(Received J a n u a r y 2, 1980; Communicated by J. B. Goodenough)

ABSTRACT The sequence of catalytic activities for the rare-earth manganites(LnMnO3, Ln=La-Eu), La > Pr = N d > S m = Eu, is correlated with the paramagnetic Weiss constant; it was positive in LaMnO3, negative in EuMnO 3.

Introduction We have studied the catalytic properties of rare-earth transition-metal mixed oxides in an attempt to obtain a gas sensor, in particular an alcohol sensor. In a preceding paper(l) we reported the catalytic properties of rare-earth copper double oxides. It was found that these properties depend on the kind of rare-earth included; drastic changes occur on going from Pr to Nd. It may be interesting, therefore, to study the effect of the rare-earth ion on the catalytic properties of other rare-earth transition-metal m i x e d oxides. In this paper we report some results obtained on the rare-earth manganites.


and Procedures

Catalyst preparation. The catalyst used was prepared by the solid-state reaction of dried Ln203 and Mn203. The well ground mixtures of components were fired at 1300°C in air for 10 hr. These compounds c o n s i s t o f a single orthorhombic phase, as determined by x-ray diffraction(Table i). Procedures. The synthesized powder(surface area;<2 m2/g) was pulverized to about 300 mesh size and mixed with a small amount of n-butyl acetate-cellulose solution to make a paste. Subsequently, 269


T. ARAKAWA, e t al.


1 X-ray


constants o

Sample LaMnO~ PrMnO~ NdMnO~ SmMnO~ EuMnO~

Symmetry orthorhombic ,, ,, ,, ,~

Vol. 15, No. 2

for LnMnO 3 o






5.533 5.445 5.406 5.357 5.335

5.722 5.786 5.789 5.825 5o826

7.692 7.574 7.556 7.482 7.485



243.6 238.7 236.5 233.5 232.3

the paste was printed on an alumina plate(8 m m x 4 m m x 1 mm) and dried. The sample was then heated at 1000°C for i0 minutes and cooled in air. Electrical contacts were made using platinum mixed paint at both ends of the painted oxide film. The film thickness was about ca. 40 ~m. The thin film was set in a pyrex glass tube, of i0 mm i.d., and nitrogen gas(contained 02 < 50 ppm) was passed through. The conductivity changes in these oxide were measured by conventional methods over the range 25-500°C; the current, which depended upon the resistance of the film, was converted to voltage and recorded by an electronic recorder. The magnetic susceptibility data for LnMnO 3 were obtained with a Shimadzu MB-II magnetic balance over the range 77-300°K. Temperature-programmed-desorption(TPD) experiments were performed as follows. The TPD apparatus used is in prin2~ ciple the same as that described in the literature(2). 7 After being mounted in the TPD O cell, each sample was degassed at 500°C for 1 hr and then exposed to oxygen(about i00 Torr) ~i~ at the same temperature. This c was cooled to 0°C under an oxygen ® atomosphere and then degassed for 15 minutes at 0°C. After these operation, the TPD cell was heated at 10°C/min. Gas 0 200 300 400 desorption during heating was Temperat u r e ( "C ) continuously monitored with mass spectroscopy. FIG. 1 !


and Discussion

The temperature dependence of the response ratio of nominal LnMnO 3 to the injection of 1 ~i methanol is shown in Fig. i. When the magnetic susceptibility of these oxides was measured before and after the reaction, the values of ~eff did not vary° Therefore, it was thought that

Response ratio vs temperature curves for CH30H adsorption. R - Ro




xl00 (%)

R: the maximum resistance after CH30H adsorption Ro: the reszstance in steady gas flow Carrier gas; N2(40 ml/min)

VoI. 15, No. 2


the state of oxidation of these samples did not change before and after the reaction. The activities of these oxides did not change throughout the experimental period. The order of the activities was La > Pr = N d > Sm = Eu, where the activity is given by the temperature at which a response ratio of 20% is attained. Thus, it is found that the activity decreases as the radius of the rare-earth ion in LnMnO 3 decreases.



LnMnO3 ;



g .a

i',; / , , z.~ Plots of inverse magnetic ,' , /, , - s' .......... susceptibilities, X~ I, against i ,' ,' J ," temperature for LnMnO3 are shown I O0 200 30O Ternperat ure('K) in Fig. 2. LaMnO3, PrMnO 3 and NdMnO 3 clearly show weak ferroFIGo 2 magnetic behavior with positive values of the paramagnetic Curie Inverse susceptibility temperatures, 172°K, 84°K and perature for LnMn03. 50°K, respectively. This result is consistent with a distortion to orthorhombic symmetry that cants spins to give a weak ferroTABLE 2 m a g n e t i s m along the c-axis, as m e n t i o n e d by Matsumoto(3). Effective magnetic moments of EuMn03 and SmMnO 3 clearly do not obey the Curie-Weiss rule, Compound ~LnMnO3 ~Ln3+ as reported by Goodenough(4). In this paper, we obtained the LaMnO. 4.60 0.00 paramagnetic Curie temperature PrMnO 3 5.96 3.56 for SmMnO 3 and EuMnO 3 by exNdMnO~ 5.79 3.34 trapolation of the linear porSmMnO 3 5.19 1.74 tion below 120°K. These EuMnO~ 5o94 3.60 temperatures are 4°K and -34°K, respectively. i

vs tem-

LnMn03 ~Mn 4.60 4.78 4.73 4.79 4.72

Analysis of the linear portion of the reciprocal of magnetic susceptibility versus T curves yields the values of ~eff presented in Table 2. It is found that the value of ~eff for the Mn ion in LaMnO 3 is somewhat smaller than the theoretical value(4.82 ~B). But it is currently not clear whether this result is due to the presence of Mn 4+ or formation of delocalized holes in a narrow Mn3+: d 4 band. For other manganites, the valence state of the Mn ion is almost trivalent since the value of ~eff for the Mn ion agrees with the theoretical value within the experimental error. Goodenough(5) has pointed out that holes in the Mn3+:d 4 m a n i f o l d produces a ferromagnetic Mn 3+- 02-- Mn 3+ double-exchange interaction in LaMn03+ x. He also pointed out that below a Jahn-Teller distortion the superexchange interactions in LaMnO 3 are ferromagnetic in two directions and antiferromagnetic in the third. A paramagnetic


T . ARAKAWA, et al.

Vol. 15, No. 2

La u 250 ==


o v




Sz.O 5 0 Mn3"- Mn3" Distance(k)


vs Nn3+-Nn 3+

interionic distance(a-axis).

Weiss constant inates. The oxygen content We have found, axis Mn3+-Mn 3% to that in the






0p FIG. 4


Nn3+-Nn 3+ i n t e r a c t i o n


The relationship between and activity. The activity i~ given by the temperautre at which a response ratio of 20% is attained.

8p>0 indicates that the ferromagnetic component dommagnitude of 0p>0 increases rapidly with increasing in LaMnO3+ x because of the double-exchange component° see Fig. 3, a linear relation between 8D and the aseparation. This result would seem to-be similar system of (La,Sr)MnO 3 obtained by Watanabe(6) o

The variation of activity with 0p is shown in Fig. 4. The activity decreases linearly with 8p, except for LaMnO3+x, which is oxidized(x>0) o Gallagher(7) has pointed out that from the thermogravimetric method the loss of oxygen in Lal_xPbxMnO 3 decreases to values as small as the cell size. From the experiment of thermal program desorption of oxygen similar to that described above, we recognized that the amount of the desorption of oxygen from LnMnO 3 is very small. Therefore, when methanol is adsorbed on LnMnO3, it would seem that a bulk oxygen of LnMnO 3 reacted with methanol and a vacancy was produced according to the following equation: Mn3+ - 02 - _Mn3 +

1/3 MeOH• Mn2+_ Vo -Mn 2+ +l/3CO 2 +2/3H20


Since the catalytic activity increases with 8~ and 8n increases with the concentration of mobile holes in ~he Mn3¢:d 4 band (via double-exchange component), our results are consistent with a catalytic activity that increases with the ease of exchanging oxygen atoms with the solid. ~+Thls, in turn, appears to correlate well with the size of the Ln j ionso The authors are happy to acknowledge the support of a Scientific Research Grant from the Ministry of Education, Japan, for part of this work.

Vol. 15, No. 2



Re ferences i) T. Arakawa, S. Takeda, Bull., 14, 507(1979).

G. Adachi and J. Shiokawa, Mat° Res.

2) M. Iwamoto, Y. Yoda, M. Egashira and T. Seiyama, 80, 1989 (1976). 3) G. Matsumoto,

IBM J. Res. Develop.,

J. Phys. Chemo,

14, 258(1970).

4) J.B. Goodenough and J.M. Longo, in " Landolt-B~rnstein" (K.H. Hellwedge Ed.), Group "liT, VOI. 4a, p 126, Springer-Verlag, Berlin (1970) . 5) J.B. Goodenough, 6) H. Watanabe,

Phys. Rev.,

i00, 564(1955).

J. Phys. Soc. Japan,

7) P.K. Gallagher, 9, 1345(1974).

D.W. Johnson,Jr.

16, 433(1961). and F. Schrey, Mat. Res. Bull.,