Bull. Mater. Sci., Vol. 27, No. 2, April 2004, pp. 179–182. © Indian Academy of Sciences.

High effective silica fume alkali activator

V
VLADIMÍR Z IVICA
Institute of Construction and Architecture of the Slovak Academy of Sciences, Dúbravská cesta 9, 845 03
Bratislava 45, Slovak Republic

MS received 11 November 2003

Abstract. Growing demands on the engineering properties of cement based materials and the urgency to
decrease unsuitable ecologic impact of Portland cement manufacturing represent significant motivation for
the development of new cement corresponding to these aspects. One category represents prospective alkali
activated cements. A significant factor influencing their properties is alkali activator used. In this paper we
present a new high effective alkali activator prepared from silica fume and its effectiveness. According to the
results obtained this activator seems to be more effective than currently used activators like natrium hydroxide,
natrium carbonate, and water glass.

Keywords. Alkali activation; slag; Portland cement; slag blends; silica fume; alumino-silicates.

1. Introduction Naturally the effectiveness of the alkali activator is a

very significant factor influencing the intensity and rate
Growing demands on the engineering properties of cement of process of alkali activation (Wang et al 1994; Ski and
based materials and ecological aspects in the manufacture Day 1996). It is reported that the best activator is water
of Portland cement are some of the motivation factors for glass (Slota 1987). Its best dosage is within the range of
the development and use in building practice of the cements 3⋅0–5⋅5% Na2O by weight of slag. The optimum dosage
corresponding to these aspects (Davidovits 1993). The ranges of water glass solution are 0⋅75–1⋅25% for acid
binding systems based on the alkali activation represent a slag, 0⋅90–1⋅30% for neutral slag, and 1⋅0–1⋅5% for basic
group of cements of this category (Davidovits 1994). slag. The optimum range of fineness is 4000–5500 cm2/g.
Alkali activated cements (AAC) are the type of the bin- It is evident that the effectiveness of activator is a very
ders synthesized from natural alumino-silicate materials significant factor which determines the engineering pro-
and alumino silicate waste products such as flyash, rice perties of the material. Therefore, the effectiveness of alkali
husk ash, slag, etc which are rich in silicium and alumi- activator represents the aim worthy of attention for fur-
nium (Herr 2002). Therefore, the use of alkali activation ther research. The subject of the paper is a new type of
technology enables the possibility of substantial utiliza- alkali activator prepared from silica fume (SFA) deve-
tion of waste materials. Moreover, it seems that AAS are loped in the author’s laboratory. Here the effectiveness of
able to enable in some respects some better engineering SFA as the activator of slag binder systems is shown.
properties opposite to those of Portland cements (Ander-
son et al 1988; Gjø′ rv 1989).
Alkali activation is a process based on the action of 2. Experimental
solutions of alkali compounds like natrium or kalium
hydroxide, natrium carbonate and silicate on the alumino- For the study of the effectiveness of SFA, mortar (20 mm)
silicate constituent of the binder. The result is the destruc- edge cubes from Portland cement (CEM I 42⋅5 according
tion of the chemical bonds of the type Si–O–Si in the to STN PEN V 197-1), their slag blends, and slag, were
aluminosilicate. prepared. Cement : sand ratio was 1 : 3 (STN 71 1208). The
Various hydration products are the result of alkali acti- fresh mortar mixtures were prepared adding SFA, natrium
vation like calcium silicate hydrates, alkali calcium alu- hydroxide and water glass. For comparison the mortar mix-
minosilicate hydrates, gehlenit and hydrogarnet hydrates. tures with water addition were prepared. The criterion of
It is known that the efficiency of an alkali activator is the dosage was the equivalent workability based on the
very strongly dependent on the physicochemical nature of test according to the standard STN 72 2441.
the material to be activated. The presence of amorphous SFA used represents a solution—a product gained from
phases, in particular from the system, CaO–Al2O3–SiO2, defined suspension of silica fume and water solution of
plays a significant role due to its considerable reactivity alkali compound. The suspension was cured under the
towards alkalies. Then, naturally towards the process of defined ambient temperature and time of action of alkali
alkali activation. solution on silica fume.
179
 v
180 Vladimír Z ivica

Table 1. Chemical composition and properties of the materials used.

Portland cement
CEM I 42⋅5 Silica fume Blast furnace
Composition (%) Mokrá Široká slag Trinec

Loss on ignition 1⋅38 2⋅35 2⋅25

Insoluble residue 1⋅18 2⋅34 –
SiO2 20⋅91 94⋅14 37⋅14
CaO 62⋅18 0⋅13 37⋅40
Al2O3 5⋅82 0⋅13 9⋅15
Fe2O3 3⋅40 0⋅18 1⋅05
MgO 2⋅36 0⋅18 11⋅70
SO3 2⋅78 0⋅28 0⋅37

Spec. weight (g⋅cm–3) 3⋅116 2⋅269 2⋅842

Spec. surface (m2⋅kg–1) 334⋅3 15898 231⋅5
Maximal size
Normal consistency 29 of grain, 90 µm
Setting:
beginning 2 h 45 min
end 4h

Mineralogical composition (Bogue) – %

C3S–42⋅46; C3A–9⋅73; CaSO4–6⋅03; C2S–27⋅90; C4AF–10⋅21

The moulds with fresh mortar test specimens were cured

for 24 h at relative humidity of ca 95%, and after demould-
ing in water at ambient temperature of 20°C.
The properties of the test specimens studied were com-
pressive strength, chemically bound water, and hydration
products produced using SDT 2960 device and standard
thermal analysis programme T.A., pore structure (total
porosity of portion macro pore with radius over 7500 nm,
pore median, surface area), using CARLO ERBA mercury
intrusion porosimeter mod 2000.
Information on the properties of the material used is
given in table 1.

3. Results and discussion

Figure 1. Development of compressive strength of mortars
3.1 Compressive strength depending on the alkali activator used and the composition of
the binder and the time of hardening. 1. Slag + NaOH; 2. Silica
The development of compressive strength of the test speci- fume activator + slag; 3. Portland cement + water; 4. 70% Port-
land cement + 30% slag + water glass; 5. 30% Portland cement
mens is shown in figure 1. It is evident that the mortars + 70% slag + water; 6. 30% Portland cement + 70% slag +
prepared by the application of SFA show higher values of silica fume activator; 7. 10% Portland cement + 90% slag
compressive strength mainly after 28 days of hardening + water; 8. 10%Portland cement + 90% slag + silica fume
opposite to those prepared with NaOH, water glass, and activator.
water. This comparison documents the higher effectivity
of SFA opposite to the effect of two mentioned activa-
after 28 days of hardening with application of SFA are
tors, and water.
clearly evident.
A comparison of the effectiveness of SFA and some
currently used activators is shown in figure 2. The high
effectiveness of SFA further confirms the comparison of 3.2 Chemically bound water and pore structure
the compressive strength values of the mortar mixtures
No. 6 and No. 8 with SFA prepared from blends contain- The results of estimation of the chemically bound water
ing 30 and 10% of Portland cement opposite to the mortar are given in table 2. For the content in chemically bound
No. 2 prepared from Portland cement and water. Signifi- water the values of loss on ignition estimated by GTA in
cantly higher values of compressive strength reached the interval 100–500°C were considered.
 High effective silica fume alkali activator 181

Figure 2. Comparison of the effectiveness of the alkali activator used. 1. Slag, silica
fume activator opposite to natrium hydroxide; 2. 30% Portland cement – 70% slag, silica
fume activator opposite to water; 3. 10% Portland cement – 90% slag, silica fume
activator opposite to water, 1 and 28 days values are equal; 4. Slag, silica fume activator
opposite to water glass.

Table 2. Composition of mortar mixtures, content in water bound and pore structure parameters after 28
days of hardening.

Composition of the mixture

Content Portion of Specific Pore
Portland L chemically Total macropores surface area median
cement Slag B bound water (%) porosity (%) r > 7500 nm (%) (m2/g) (nm)

30 70 0⋅52 1⋅37 18⋅9 21⋅06 5⋅0 19

water
30 70 0⋅74 1⋅01 10⋅9 24⋅00 4⋅2 11
SFA
10 90 0⋅52 3⋅30 20⋅1 17⋅08 3⋅3 20
water
10 90 0⋅74 3⋅30 12⋅5 27⋅55 3⋅3 13
SFA
– 100 0⋅42 2⋅69 23⋅8 38⋅34 2⋅7 24
water
7% Na2O
– 100 0⋅80 1⋅43 16⋅8 33⋅70 1⋅4 17
SFA

L
, ratio of dosage of water or SFA and binder.
B
 v
182 Vladimír Z ivica

According to the results obtained, it may be seen that it significantly overcomes the current alkali activators like
the content in chemically bound water or the hydration natrium hydroxide and water glass. Therefore, silica fume
products, respectively was in the mortars with application activator seems to be worthy of further attention in the
of SFA lower or equal to this estimated in the mortars interest of further increase of its effectiveness. The follow-
prepared with water. ing topics of research seem to be interesting and useful:
The values of pore structure parameter also given in
table 2 show the lower values of total porosity and pore – the dependence of effectiveness of activator on the che-
median at the SFA mortars opposite to those prepared with mical and mineralogical composition of the material to be
water. Opposite relationship in the values of macropore activated,
portion and surface area may be observed. The decreased – the possibility of elimination of the found negative
total porosity indicated an increased content of hydration effect of the silica fume activator application on the deve-
products in the SFA mortars. The decreased content of lopment of pore structure of the activated material occur-
chemically bound water found indicates that the hydra- ring as an increase of the content in macropores.
tion products produced under the application of SFA were
the compounds with decreased content in chemically bound
water. Therefore, even the content in chemically bound Acknowledgement
water should be higher. But it was found that this was
lower. Therefore, it may be assumed that the hydration The author would like to thank the Slovak Grant Agency,
products produced under the application of SFA are com- VEGA, for its support of this work (grant 2/3039/23).
pounds having lower content in the chemically bound
water as those produced under the use of water.
The decreased pore median values found indicate that References
the hydration produced under the application of SFA may
formate finer pore structure than those produced at the Anderson A, Gram H E, Malolepszy J and Dejy J 1988 Alkali
use of water. Both the effects—porosity and pore median activated slag (Stockholm: Swedish Cement and Concrete
decrease—are known as factors contributing to the posi- Research Institute) p. 104
tive development and strength increase of cement based Davidovits J 1993 Ceram. Trans. 37 165
materials. It was evident from SFA application. Davidovits J 1994 High alkali cements for 21st century con-
The results of both positive pore structure effects of cretes, Concrete technology past, present and future, V. Mohan
Malhotra Symposium (ed.) P Kumar Mehta (Detroit: Ameri-
SFA application may be, however, limited in some extent
can Concrete Institute) pp 383–397
by the adverse effects contributing to the decrease of com-
Gjø′ rv O E 1989 Alkali activation of Norwegian granula-
pressive strength. These represent increased values of ted blast furnace slag, 3rd int. conf. on flyash, silica fume,
macropore portion, and decreased values of specific sur- slag and natural pozzolan in concrete (ed.) V M Malhotra
face area of SFA mortars opposite to those at the use of (Detroit: American Concrete Institute), Vol. 2, pp 1501–
water (table 2). Both considered effects are known as fac- 1517
tors influencing slowing down of strength development Herr R 2002 Die Entwicklung von Baustoffen nach dem Prinzip
and their decrease. der Geopolymerization, IEMB, Neue Ergebnise 2002, 6 ps
What is geopolymer? www.geopolymer.org/about3.html
4. Conclusions Ski Caijun and Day R L 1996 Cem. & Concr. Res. 26 439
Slota R J 1987 Cem. & Concr. Res. 17 703
The results obtained show high alkali activation effec- Wang Shao-Dong, Scrivener K L and Pratt P L 1994 Cem. &
tiveness of new silica fume activator. As it has been shown Concr. Res. 24 1033