ABSTRACT - IPEN



ELECTRICAL CHARACTERIZATION OF TIN DIOXIDE BASED CERAMICS

DOPED WITH Mn, Sb AND Cr

W. Lacerda Jr*, W. C. Las, M. Cilense, M. A. Zaghete

*R. Professor Francisco Degni, S/N(, CEP 14801-970 - Araraquara-SP

Wil@iq.unesp.br

Instituto de Química – UNESP – Araraquara – SP

ABSTRACT

SnO2-based ceramics were investigated as to their electrical properties, specifically those related to the non-linear behavior found in varistor devices. SnO2 systems were prepared containing additives to promote densification (Mn or Cu), electrical conductivity (Sb) and the varistor effect (Cr). Surface area measurements and X-ray diffraction analysis were used to characterize the precursor powders. Electrical measurements ( I-V plots) of the sintered compacts were carried out and the non-linear behavior was observed for all studied systems. Copper doped systems are more conductive and have lower non-linear coefficients and breakdown fields than those doped with manganese. Increasing antimony concentration yields a more conductive ceramics. The effect of Cr addition (0.05% mol) was small and different for systems containing Mn and Cu. The Sn-0.7%Mn-0.15%Sb-0.05%Cr system presented the highest non-linear coefficient value, (, around 13. Therefore, the main non-linear effect observed in the studied ceramics comes mainly from manganese and antimony.

Keywords: tin dioxide, varistor, manganese, antimony, chromium.

INTRODUCTION

Varistors are electronic devices with the prime function of feeling and limiting transient voltages repeatedly without self-destruction. They present non-linear I-V characteristics, similar to Zener diodes. But, differently to the Zener diode, varistors can limit overvoltages equally at both polarities. Varistors can be used in alternated current (a.c.) or direct current (d.c.), from microamperes to kiloamperes. They can also be used, in a large range of voltages, from few volts to several kilovolts. Another important varistor property is the great capacity for energy absorption, from few joules to hundreds of joules. Because of their versatility, varistors have been used in electro-electronic and semiconductor industries.

The non-ohmic behavior is expressed, quantitatively, by the non-linear coefficient, (, taken from the expression [pic], where J is the current density, E is the electric field and k is a constant which depends on material microstructure. For a non-linear behavior, thus, ( must be greater than 1, so that, the greater the ( value the better the varistor(1). Another parameter used to characterize the varistor behavior is the breakdown field, ER, defined as the field where deviation from linearity starts. The J x E curves are derived from I-V measurements, so the term I-V characteristics will be employed.

The parameter ( is influenced by type and concentration of dopants in the system. Bueno et al(2) studied the ceramics (98.95%SnO2 + 1.0%CoO + 0.05%Nb2O5) prepared by conventional route, obtaining ( = 8 and breakdown field ER = 1870 V(cm-1. When 0.05% of Cr2O3 was added to the system, the values of ( and ER increased to 41 and 3990 V(cm-1, respectively. The addition of higher Cr concentrations caused a deleterious effect in the electrical properties, in such a way that the varistor resistance increased four orders of magnitude and the non-linear behavior was not observed. Probably, ER was higher than the apparatus limit of detection. In the present work, results for SnO2-based systems prepared by organic route are presented.

MATERIALS AND METHODS

Four systems were prepared: 1 - (Sn - 0.7%Cu - 0.15%Sb); 2 - (Sn - 0.7%Mn - 0.15%Sb); 3 - (Sn - 0.7%Cu - 0.15%Sb - 0.05%Cr) and 4 - (Sn - 0.7%Mn - 0.15%Sb - 0.05%Cr). Following the procedure developed by Pechini(3), a standardized tin solution was weighed to yield 20 g of SnO2. Next, this mass of solution was doped with the Pechini solutions of the respective dopants with their mass previously calculated for the required doping. These mixed solutions were concentrated into a polymeric black mass and treated at 400(C for 4 h to eliminate the organic portion. Next, powders were crushed in a mortar, calcined at 500(C for 15 h and analysed by X-ray diffraction to verify the formation of the SnO2 phase. Powders were pressed uniaxially using a 8 mm die and isostatically at 210 MPa. Pressed compacts were sintered in a tube furnace LINDBERG with MoSi2 resistance. Ceramics containing Cu were sintered at 1250(C for 4 h and those containing Mn were sintered at 1300(C for 4 h, because systems with Cu densify at lower temperatures than those with Mn. I-V measurements were performed at room temperature and normal atmosphere by using a high voltage source-measure unit Keithley, model 237, (internal impedance: 1014 (), on pulse mode (100 ms width pulse) and on continuous mode, from 0 to 1100 V. Ohmic contacts were made by applying silver paint on parallel surface ceramics, which were cured at 300(C for 5 h.

RESULTS AND DISCUSSION

Figure 1 shows the comparison of Cu and Mn systems without Cr. We can observe that Cu systems are more conductive and have lower ( and ER values than Mn ones (see also Table I discussed at the end of this section), even with different sintering temperatures to compensate for possible differences in electrical properties due to different densification rates. Grain size of systems containing Cu and Mn are very close to each other, around 5 (m and 7 (m, respectively. Thus, the large difference in conductivity cannot be related to grain size, but, perhaps, to the presence of higher quantities of Sb in the grain boundary region.

[pic]

FIGURE 1 - I-V characteristics (d.c.): (a) and (b) Sn-0.7%Mn-0.15%Sb,

(c) and (d) Sn-0.7%Cu-0.15%Sb.

It is known that CuSb2O6 and MnSb2O6 compounds are formed, respectivily, in the 600-700(C and 950-1050(C temperature ranges(4). Depending if these compounds segregate on grain surfaces and depending on their relative resistivity, it would be possible to explain the higher conductivity of the copper system. The fact that antimony segregates in the presence of copper and not of niobium or tantalum has already been reported(5). What happens to manganese is not known. Until now, these phases could not be detected on SnO2-based ceramics.

Figure 2 shows the results of Cu doped ceramics with and without Cr. Two discs of each system were measured, showing there is reproducibility in ceramics preparation and electrical characterization. This is verified for all measurements throughout the work. Cr addition produced a slightly more conductive ceramics, with ( value similar to the one found for ceramics without Cr.

[pic]

FIGURE 2 - I-V characteristics (d.c.): (a) and (b) Sn-0.7%Cu-0.15%Sb;

(c) and (d) Sn-0.7%Cu-0.15%Sb-0.05%Cr.

Figure 3 shows the I-V curves for ceramics with Mn, checking the influence of Cr presence. We observe that the influence of Cr is to increase somewhat the resistivity, without changing much the ( value, as seen in Table I. On the other hand, the breakdown field increases with Cr addition. It seems the role of Cr is different in Mn and Cu doped ceramics, but in order to verify the extent of its influence other studies varying Cr concentration are needed.

[pic]

FIGURE 3 - I-V characteristics (d.c.): (a) and (b) Sn-0.7%Mn-0.15%Sb,

(c) and (d) Sn-0.7%Mn-0.15%Sb-0.05%Cr.

Figure 4 shows the Sb concentration effect on I-V curves. An increase in Sb concentration increases conductivity by more than two orders of magnitude and brings the ER parameter to lower field values. There are two possibilities: either the system with lower Sb content will show the breakdown region, but at fields much higher than those within the equipment working range, or it will be always linear.

[pic]

FIGURE 4 - I-V characteristics (d.c.): (a) Sn-0.7%Mn-0.01%Sb-0.05%Cr;

(b) Sn-0.7%Mn-0.15%Sb-0.05%Cr.

In the second case, there would be an optimum Sb concentration for which the ( value would be higher, since it is known that for higher Sb content the ceramics becomes conductive and ohmic(6). For the time being, the equipment used indicates that the system with 0.15% Sb shows a higher ( value.

Table I shows ( and ER values obtained for 0.15%Sb sintered ceramics. Systems with Cu were not measured in the pulse mode for experimental problems.

Table I: Non-linear coefficients for SnO2-based systems

|SYSTEMS |( (d.c.) |( (pulse mode) |ER / V(cm-1 |

|Sn-0.7%Cu-0.15%Sb |02.3 |... |0305 |

| |03.1 |... | |

|Sn-0.7%Cu-0.15%Sb-0.05%Cr |02.5 |... |0289 |

| |02.8 |... | |

|Sn-0.7%Mn-0.15%Sb |11.1 |5.9 |0941 |

| |11.2 |6.4 | |

|Sn-0.7%Mn-0.15%Sb-0.05%Cr |11.5 |7.7 |1256 |

| |13.6 |7.1 | |

Systems with Mn have ( values higher than 10, while those with Cu have around 3, showing that Mn is more effective than Cu in obtaining the non-linear characteristics, at least for the sintering and measuring conditions used. Also, Mn doped ceramics have higher breakdown fields. On the other hand, in the quantity used, Cr does not have a significative effect in the ( value. Systems containing Mn are more resistive, which allowed measurements on pulse mode. The higher ( values obtained for d.c. mode than for pulse mode were caused by the fact that the ceramics experienced more heating on d.c. mode. This is so, because ceramics heating increases the electrical conductivity, causing an increase in current passing through the varistor. This augmented current increases the ( values as compared to those obtained for pulse mode, where no heating is expected. All ( values were obtained by linear fit in the region of elevated applied electric field.

CONCLUSION

SnO2-based systems containing Mn presented higher non-linear coefficients, breakdown fields and resistivities than those containing Cu, probably because of different Mn and Cu segregation associated with Sb. The effect of Cr was small and perhaps different in both systems, with Mn and Cu, but needs further investigation. Increasing Sb content in Mn doped ceramics renders a more conductive material, brings the breakdown field to lower values and could indicate an optimum concentration for the non-linear behavior. The best system obtained in the present study is Sn-0.7%Mn-0.15%Sb-0.05%Cr, with (=13 and ER =1256 V/cm related rather to Mn and Sb presence than to Cr.

ACKNOWLEDGEMENTS

The authors thank FAPESP and FINEP/PADCT for financial support.

REFERENCES

1. Gupta, T. K. - "Application of zinc oxide varistors". J. Am. Ceram. Soc., 73 [7] 1817-1840, 1990.

2. Bueno, P. R; Pereira, E. C; L. O. S., Bulhões; E. Longo; Pianaro, S. A.; Varela, J. A.-" Um modelo para descrever o mecanismo de formação da barreira de potencial no contorno de grão de varistores à base de dióxido de estanho". Cerâmica, 42 (277), 618 - 621, 1996.

3. Pechini, M.P., “Method of preparing lead and alkaline earth titanates and niobates and coating method using the same to form a capacitor”, U.S. Patent nº 3330697, 11 julho 1967; LESSING , P.A., “Mixed-cation oxide powders via polymeric precursors”, Ceramic Bull., 68 (5), 1002-1007, 1989.

4. Grigoryan, L. T., Gedakyan, Dzh., Kostanyan, K. A. “Activated sintering of stannic oxide”, Inorg. Mat., v. 12, p. 313-314, 1976.

5. Las, W.C., Dolet, N., Dordor, P. and Bonnet, J.P. “Influence of additives on the electrical properties of dense SnO2 -based ceramics”, J. Appl. Phys., 74(10), 6191-6196, 1993.

6. Mazali, I.O., "Efeito do método de preparação e da concentração de antimônio na sinterização e propriedades elétricas de cerâmicas densas à base de dióxido de estanho", Tese de Mestrado, IQ - UNESP, Araraquara - SP, 1997.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download