Inorganic polymers made of fayalite slag: On the microstructure and behavior of Fe

Abstract

The microstructure of inorganic polymers (IP) formed from fayalite slag was investigated as a function of the composition of different activating solutions. The starting slag was 80 wt% amorphous, and after activation using sodium silicate solutions with varying SiO2/Na2O molar ratios, the amorphous phase dissolved and a binder phase was formed. The morphology of this binder, including the population and size of remnant particles and pores, was dependent on the particular activating solution used, and became denser as the level of silicate rose. 57Fe Mössbauer spectroscopy revealed that the IP synthesis reaction is combined with the oxidation of Fe2+ from the fayalite slag to Fe3+ in the inorganic polymer binder. The reaction extent varied and could be quantified using the absorption areas of these ions. Data corroborate that the Fe2+ ions in the amorphous part of the fayalite slag and the Fe3+ ions in the new binder phase had an average oxygen‐coordination number of 5.

Molecular structure of CaO–FeOx–SiO2 glassy slags and resultant inorganic polymer binders

“…However, these results are approximate since a wide range in values can be obtained from the calculation, as explained further in the supplementary information. Previous work using XANES and Mössbauer spectroscopy showed the presence of IV Fe 2+ , V Fe 2+ and a minor amount of V Fe 3+ species in a similar slag, which, in comparison with literature, is in agreement with the Fe–O peak found in Figure at 2.01‐2.02 Å.…”

“…3.6 Å . In contrast to these AAMs, the Fe‐rich IPs only show subtle atomistic changes due to the alkali‐activation reaction, which is in agreement with the observed changes seen via Mössbauer spectroscopy, where the parameters, eg, isomer shift and quadrupole split, of the Fe‐species of the slag and IP are related …”

“…This is partially due to the presence of unreacted slag residing in the binders, where approx. 30‐60 wt.% of unreacted slag has been observed in similar systems . Based on the activating solutions used in this current study (resulting in a similar Na/Fe), the unreacted portion should be closer to 60 wt.%.…”

“…Interestingly, recent results have shown that having Fe 2+ in the precursor slag is actually preferred over Fe 3+ , leading to faster reaction kinetics and higher compressive strengths (L. Arnout, A. Peys, L. Machiels, P. T. Jones, S. Arnout, E. Nagels, B. Blanpain & Y. Pontikes, unpublished data). Onisei et al confirmed that the Fe‐rich IP binder consists of Fe 3+ using titrations and Mössbauer spectroscopy, respectively. Furthermore, the use of synchrotron X‐ray absorption near‐edge spectroscopy (XANES) has given better insight into the neighboring environment of the Fe atom, however, the average coordination number of the Fe atoms remains an area of debate in the literature, as Mössbauer results suggest an average number of 4‐5, while XANES implies that 5‐5.5 is more reasonable .…”

“…Onisei et al confirmed that the Fe‐rich IP binder consists of Fe 3+ using titrations and Mössbauer spectroscopy, respectively. Furthermore, the use of synchrotron X‐ray absorption near‐edge spectroscopy (XANES) has given better insight into the neighboring environment of the Fe atom, however, the average coordination number of the Fe atoms remains an area of debate in the literature, as Mössbauer results suggest an average number of 4‐5, while XANES implies that 5‐5.5 is more reasonable . Taking into account both the results from Mössbauer and XANES, an average 5‐fold coordinated iron is the most probable.…”