ELECTRICAL CHARACTERIZATION OF A FAST-SETTING GEOPOLYMER

Elder A. de Vasconcelos(1), Erika P. Marinho(1),Saulo F. Oliveira(1)

and Eronides F. da Silva Jr.(2).

(1) Universidade Federal de Pernambuco. Centro Acadêmico do Agreste. Rodovia BR-104 km 59. Nova Caruaru, 55002-970, Caruaru-PE.

email:

(2) Universidade Federal de Pernambuco.Departamento de Física.

ABSTRACT

The complex impedance spectrum of a 3.5:1 (SiO2:Al2O3) structural geopolymer prepared using metakaolin as geopolymeric precursor shows two arc sections, similar to the spectra observed in ordinary Portland materials. This similarity can be understood considering that its overall electrical response is due to 3 different conduction pathways in the material bulk plus effects at the electrode/material interface.An equivalent circuit to account for the spectrum was presented.

Key-words:Geopolymers, impedance spectroscopy, cement.

1. INTRODUCTION

Geopolymers are inorganic polymers produced by room temperature polymerization reactions of aluminosilicates or silicates in strongly alkaline environments. They have potential to be used in a variety of applications, including fast-setting concrete and toxic or radioactive waste storage, for example. In a previous work(1), we showed that a 3.5:1 (SiO2:Al2O3) structural geopolymer prepared using metakaolin as geopolymeric precursor in NaOH solution, hereafter identified as GPMK, settled rather quickly, in approximately 20 min, but had a mechanical strength compatible to values typical of structural materials. The compressivestrength measured after 24 hwas 4.4 MPa.

In general, so far, there have been are few reports on the electrical properties of geopolymers. In this work, we report on the impedance of GPMK. It will be shown that its impedance spectrum issimilar to the spectra observed in ordinary Portland materials. The reasons for this similarity will be also discussed.

2. EXPERIMENTAL DETAILS

The GPMK samples were synthesized following a procedure describe previously1. Theadditional step in this work was to introduce 2 clean copper electrodes (4.5 cm x 4.5 cm) in the slurry in the molds (5 cm x 5 cm x 15 cm) before curing. The electrodes were 14 cm apart and were contacted by means of alligator clips to an Agilent 4284A Precision LCR Meter. The impedance spectra were measured from100 Hz to 1 MHz using a test signal of 50 mV without applied bias, using both open and short circuit corrections.

3. RESULTS AND DISCUSSION

The complex impedance plot displayed in Fig. 1 shows two arc sections, with a larger arc forming the right-hand side and a smaller arc forming the left-hand side. Both arcs are not fully developed due to the upper and lower frequency limits in our measurements.

Geopolymer structure and chemistry differs in several aspects from ordinary Portland cement chemistry. Nevertheless, the impedance spectrum in Fig. 1 is remarkably similar to the spectra observed in ordinary Portland materials (cement pastes, mortar and concrete). This is not unexpected, as we will now discuss.

Fig. 1. Complex impedance plot of a GMPK sample.

Despite the specific differences in the microstructure of geopolymers and ordinary Portland cimentitious materials, their overall electrical response can be understood from a general viewpoint. When an electric field is applied across the material, current flows by a number of different conduction pathways, which can be classified into three kinds: (1) low resistance, continuous paths; (2) discontinuous conductive paths separated by insulating barriers and (3) high-resistance paths(2).For example, hydrated cement paste is frequently viewed as an insulating matrix containing a micropore network filled with a conductive electrolyte. Electrical conduction through paste samples is made possible by the net movement of charge carrying ions such as Ca++, Na+, OH- and SO4= in the electrolyte. On the other hand, GMPK is, ona nanometer scale, a complex arrangement consisting of a porous network of clusters of silico-aluminate structures containing ions(3).In other words, it is an insulating porous matrix containing ions. Therefore, one can expect that these 3 kinds of paths might be also present in GPMK.

Frequently, itis convenient to separate the electrical response into two distinct regions, one relating to electrode/material interface processes and the other due to processes in the material bulk. Usually, resistors are used to describe irreversible processes which dissipate energy, (charge transport and recombination across an interface, for example) while capacitors are used to describe reversible processes (charge trapping at interface dangling bonds, for example). The electrode/material interface effects are usually accounted for by a parallel RC circuit. Electrode effects should be considered carefully, otherwise the interpretation of bulk process can be obscured(4,5). Fig. 2 shows the resulting equivalent circuit. It is a very general model, with potential to describe successfully the electrical response related with a variety of phenomena, including, cement hydration(2,6)and chloride transport in concrete mixtures(7). So far, our results indicate that it is also appropriate for GPMK.As Fig. 1 shows, a typical two-arc response is observed, the right-hand side is related to the electrodes and the left-hand sideis related to conduction processes in the bulk.

Fig. 2. Equivalent circuit model for the GMPK samples.

4. CONCLUSIONS

The complex impedance spectrum of GMPK was similar to the spectra observed in ordinary Portland materials: a characteristic two-arc response where the right-hand arc is related to the electrodes and the left-hand side is related to bulk. Both arcs werenot fully developed due to the upper and lower frequency limits in the measurements. This similarity indicates that the overall electrical response can be understood considering 3 different conduction pathways in the bulk plus effects at the electrode/material interface, such as charge transport and trapping.

ACKNOWLEDGEMENTS

This work was supported by grants 478128/2010-0 and 564739/2010-3 from CNPq (Conselho Nacional de Pesquisas), APQ-0898-3.03/10 and BIC-0856-1.05/12 from FACEPE (Fundação de Amparo à Ciência e Tecnologia do Estado de Pernambuco).

REFERENCES

1. MARTINELLI, A. E.;MELO, D. M. A.;MARINHO, E. P.;BATISTA, T. W. J.;ARAUJO, R. G. S.;MELO, M. A. F. Synthesis, rheological behavior and mechanical characterization of structural fast-setting geopolymers.Materials Science Forum, v.498-499,p. 488-493, 2005.

2. SONG, G. Equivalent circuit model for AC electrochemical impedancespectroscopy of concrete.Cement and Concrete Research, v. 30, p.1723-1730, 2000.

3. DUXSON, P.; FERNÁNDEZ-JIMENÉZ, A.; PROVIS, J. L.; LUKEY, G. C.; PALOMO, A.; VAN DEVENTER, J. S. J. Geopolymer technology: the current state of the art. Journal of Materials Science, v.42, p. 2917-2933, 2007.

4. FORD, S. J.; MASON, T. O.; CHRISTENSEN, B. J.; COVERDALE, R. T.; JENNINGS, H. M.; GARBOCZI, E. J. Electrode configurations and impedance spectra of cement pastes. Journal of Materials Science, v. 30, p. 1217-1224, 1995.

5. FORD, S. J.; SHANE, J. D.; MASON, T. O. Assignement of features in impedance spectra of the cement-paste/steel system. Cement and Concrete Research, v. 28, p. 1737-1751, 1998.

6. CABEZA, M.; MERINO, P.; MIRANDA, A.; NÓVOA, X. R.; SANCHEZ, I. Impedance spectroscopy study of hardened Portland cement paste. Cement and Concrete Research, v. 32, p. 881-891, 2002.

7. NEITHALATH, N.; JAIN, J. Relating rapid chloride transport parameters of concretes to microstructural features. Cement and Concrete Research, v. 40, p. 1041-1051, 2010.