Alkaline Activation of Synthetic Gehlenite Glasses: by J. Deja
Alkaline Activation of Synthetic Gehlenite Glasses: by J. Deja
by J. Deja
10 Deja
Jan Deja, D.Sc. (Eng.) is working at the AGH University of Science and Technology,
Kraków, Poland. His special field is technology of binding materials and concrete. He is
concentrated on the utilization of secondary materials in binders and concrete production.
Alkali-activated slag binders has been his subject for over 20 years.
INTRODUCTION
The use of by-product and waste materials plays also a very important role in solving
present ecological problems (for example: reduction of CO2 emission) and could enlarge
the range of building materials and partly replace high energy–consuming portland cement
[1-7].
Alkaline activation of aluminosilicate glasses is one of the new methods of special low-energy
binders production. Ground granulated blast furnace slag is the main material which is used in
this technology, but other glassy materials can be also activated (for example: fly ash,
incineration slags). Sodium carbonate, water glass or sodium hydroxide are typical alkaline
activators. Depending on the initial raw materials used, the type of activator and the conditions
of curing, the products of hydration in such binding systems can be minerals, such as C-S-H
and C-A-S-H in very dense and amorphous form. Changes of A/S and C/S ratios in these
systems make it possible to form calcium-alumino-silica gel (the main product of so-called
geopolymers), which can be good precursor for zeolite formation. High strength, very high
durability in aggressive solutions and very dense microstructure are the most interesting
properties of these materials. It gives many opportunities for their application (immobilization
of heavy metals and toxic substances, anti-filtration screens etc.) [8-17].
The objective of the study consisted in investigating the properties of alkali-activated
pastes based on synthetic gehlenite glasses. Influence of activator content and curing
conditions on the hydration process was determined. Mechanical properties as well as
heat of hydration of the pastes were measured. The phase composition and microstructure
observations of the hydrated pastes were done.
EXPERIMENTAL
Heat of hydration
The heat evolution on hydration of pastes prepared at water to solid ratio 0.5 was
followed within the 24h by microcalorimetry (BMR differential microcalorimeter
constructed in the Institute of Physical Chemistry, Polish Academy of Science). 1.2, 3.5
and 7.0% Na2O was added as activator. The initial temperature was kept constant at 25oC.
The calculated heat values are presented in Table 4.
On the heat evolution curves of the pastes, in all cases a strong initial peak appears within
the first few minutes. It is commonly accepted that very quick hydration of slags
activated by sodium hydroxide is an effect of high concentration of OH-, which can break
the bonds inside the glass structure [18]. It is clear that NaOH activated synthetic glasses
do not show an appreciable induction period. Total heat evolution during 24 hours of
hydration is very low (89 - 141 J/g) and typical for alkali-activated slag binders.
CONCLUSIONS
On the basis of the results of the research the following conclusions can be made:
• Alkaline activation of gehlenite slag glasses is influenced by the Al2O3/SiO2 molar
ratio of the slag and concentration of Na2O.
• The hydration process is much quicker in the case of gehlenite type glasses than for
typical industrial melilite granulated blast furnace slags. Total heat of hydration
measured during 24 hours of the process is much lower than for typical portland
cements.
• The highest strength has been found for pastes based on gehlenite type glass with
alumina to silica ratio 1.0.
• The results of XRD and SEM/EDS examinations show that in gehlenite type pastes
amorphous C-A-S-H phase containing high amount of sodium, silicon and aluminium
are dominating hydration product.
• 29Si MAS-NMR spectra shows that ordering corresponding to Q2 structure with
thetraedra rich in Al is dominating in the hydration products. This may be explained by
the increase in the chain-like quantity of the product, and also by the increase in the
length of chains.
• Calcium-alumino-silica gel observed in gehlenite pastes can be good precursor for
zeolite formation. SEM and XRD examinations of autoclaved samples with A/S ratio
1.2 confirmed this.
ACKNOWLEDGEMENTS
The research reported in this paper was carried out within the limits of the projects No. 4
T08 D 027 23 and No. 5 T12A 026 25 sponsored by Polish State Committee for
Scientific Research.
14 Deja
REFERENCES
Fig. 2—XRD patterns of the pastes cured at 20°C. Notations: CASH = C-A-S-H; A =
aragonite (CaCO3), C = calcite (CaCO3), S = stratlingite (Ca2Al2SiO7×8H2O).
Fly Ash, Silica Fume, Slag, and Natural Pozzolans in Concrete 17
Fig. 3—XRD patterns of the AN3 pastes (cured at 20 °C and autoclaved). Notations:
C = calcite; A = aragonite; S = stratlingite (Ca2Al2SiO7×8H2O); Q = quartz;
T = tobermorite; G = gehlenite.
Fig. 5—SEM-EDS from slag glass paste A – typical microstructure of sample AN3
(magnification 1,000 and 3,500x).
Fly Ash, Silica Fume, Slag, and Natural Pozzolans in Concrete 19
Fig. 6—Microstructure of hydration products and EDS analyse of AN3 paste after
autoclaving (magnification 2,000x).
20 Deja
Fig. 7—SEM-EDS from slag glass paste BN3 (curing at 20°C) – typical microstructure
(magnification 1,500x).
Fly Ash, Silica Fume, Slag, and Natural Pozzolans in Concrete 21
Fig. 8—Hydration products and EDS analyse of BN3 paste after autoclaving
(magnification 1,000 and 3,500x).
22 Deja
Fig. 9—29SiNMR spectra for syntetic gehlenite glass A and paste AN3
(cured at 20°C and autoclaved).
Fig. 10—29SiNMR spectra for syntetic gehlenite glass B and paste BN3 (cured at 20 °C
and autoclaved).