Content uploaded by Igor Rudenko
Author content
All content in this area was uploaded by Igor Rudenko on Jan 23, 2019
Content may be subject to copyright.
2nd International RILEM/COST Conference on Early Age Cracking and
Serviceability in Cement-based Materials and Structures - EAC2
12–14 September 2017, ULB-VUB, Brussels, Belgium
*pavlo.kryvenko@gmail.com; phone +380 (44) 2454830
CONTROL OF EARLY AGE CRACKING IN
EARLY-STRENGTH CONCRETE
BASED ON ALKALI-ACTIVATED SLAG CEMENT
Pavel Krivenko *a, Oleg Petropavlovskiy a, Igor Rudenko a, Serhii Lakusta a
a Scientific Research Institute for Binders and Materials,
Kyiv National University of Construction and Architecture, Kyiv, Ukraine
ABSTRACT
Formation of properties of high performance concretes with good durability, not depending upon cement type used, is
connected with a possibility to regulate phase composition of hydration products, pore structure and morphology of
cement stone structural units at the early stages of hydration. A specific feature of the alkali-activated cements is the
formation of low basic calcium silicate hydrates in a gel-like state. This is a reason of their sensibility to cracking at early
ages due to shrinkage. The semi-adiabatic calorimetric method was used to study early structure formation of super quick
hardening alkali-activated slag cements made using high modulus sodium silicate hydrate as alkaline component. The
alkali-activated slag cements can attain values of the compressive strength 15 MPa after 3 h and up to values 80 MPa
at the age of 28 days. Effective types of modifiers have been chosen and tried to control hydration and early age structure
formation of a cement stone to provide the lower heat evolution of these cements during hydration and the formation of a
highly homogeneous and finely porous structure of the resulted cement matrix. The results of tests showed the enhanced
performance properties of fine aggregate concretes based on alkali-activated slag cements.
Keywords: alkali-activated cement, heat evolution, structure, properties, cracking
1. INTRODUCTION
Alkali-activated slag cements1,2,3 (further, the AASCs) are characterized by a complex and unique microstructure4 due to
the absence of high-basic hydration products, rigid skeleton from portlandite and calcium aluminates, structure with well
developed surface, and the formation of mainly low- basic silicate hydrates and alkaline and alkaline-alkaline-earth
aluminium silicate hydrates. These characteristic features of the AASCs explain high strength, increased isomorphism of
strength under compressive and bending loads, high adhesion properties and durability4,5. Of special attention are the
AASCs for rapid restoration of damaged portland cement concrete. One reason is their high performance properties, the
other one is great money saving compared to traditional cements6. One of the disadvantages of the AASCs is the higher
shrinkage compared to Portland cement, especially in case of using sodium silicate hydrate as alkaline component 4,6,7,8.
This has effect on crack resistance and durability of the concretes based on such cements. This disadvantage can be
eliminated by control of early age structure formation of the AASC pastes and by design of concrete mixture with
minimum deformations of the AASC matrix.
The purpose of the study was to develop methods for control of early age structure formation of quick-hardening AASCs
to reduce early age cracking as one of the factors determining durability of high early-strength concretes.
2. RAW MATERIALS AND TESTING TECHNIQUES
The cement Type I AASC as classified in the national standard of Ukraine DSTU B. V.2.7-181 was used. Sodium
silicate hydrate (silicate modulus = 2.8, density = 1380 kg/m3) was used as alkaline component. Ground granulated blast
furnace slag (basicity modulus = 1.1, content of glass phase = 84 %, fineness measured as specific surface by Blaine =
442 m2/kg) (further, the slag) was used as aluminosilicate component. The chemical composition of the slag was, % by
mass: SiO2 – 37.90, Al2O3 – 6.85, Fe2O3 – 0.31, MnO – 0.11, MgO – 5.21, CaO – 44.60, R2O – 1.13, TiO2 – 0.35.
In accordance with the known results6,7,8, Na3PO4.12H2O salt was used as a modifier to increase resistance of sodium
silicate hydrate to coagulation upon interaction with the slag. The salt content was determined experimentally until the
reasonable initial setting time by the method of EN 196-3 had been achieved (at least 15 – 20 min as for thee case of high
early-strength cements)9. This salt was preliminary dissolved in sodium silicate hydrate and the density of the resulting
homogeneous solution was 1380 kg/m3.
2nd International RILEM/COST Conference on Early Age Cracking and
Serviceability in Cement-based Materials and Structures - EAC2
12–14 September 2017, ULB-VUB, Brussels, Belgium
Trihydric polyol (glycerin) was also used as a modifier for sodium silicate hydrate as being so-called crosslinking
admixtures for sodium silicate hydrate xerogel structures with the increased strength and water resistance10. Additionally,
glycerin is known to accelerate slag hydration in the presence of sodium silicate hydrate with formation of the additional
structural joints in a cement stone11. This predetermines the increased strength of a cement matrix under bending loads.
The application of this modifier allowed to propose the AASC system “slag - sodium silicate hydrate – glycerin” for the
mortars and concretes intended for rapid restoration of various concrete surfaces11. However, the authors did not consider
the effect of this modifier on structure formation of the AASCs and only declared high performance properties of the
mortars and concretes based on these cements.
Hydrated lime (calcium hydroxide powder Ca(OH)2) was additionally used to accelerate formation of calcium silicate
hydrates at the initial stages of AASC hardening.
River sand (fineness modulus = 1.2), as well as mixture of this sand and slag (fraction 0.315-0.63 mm) were used as fine
aggregates in the preparation of the AASC mortars.
Calorimetry of hydration in combination with kinetics of cement hardening allows for to characterize sufficiently initial
processes of structure formation from the point of view of easiness-in-production and use, strength, stress-strain state of
mortars and concretes 2,13. This was taken into account in choice of appropriate calorimetric method.
Hydration heat of the AASCs was measured on cement pastes as prescribed by the method of EN 196-9, known as
Langavant method, with some modifications. A semi-adiabatic calorimeter of own design was used in the study. Initial
temperature of the AASC paste constituents was 20 ± 1 oC. The data were continuously recorded. The interval between
measurements during hydration was not shorter than 15 s and no longer than 5 min. The total hydration heat was
computed by integrating the area under the rate evolution curve. Preparation of the AASC pastes was done as per EN
196-3. A ratio between the modified sodium silicate hydrate and slag was taken equal to 0.5 and was assumed to be
constant in the AASC pastes. In the AASC mortars, this ratio was chosen experimentally to provide flow values of 115 to
140 mm on a standard jolting table.
Drying shrinkage of the AASC mortars (fine aggregate concretes, cement/ sand = 1/3) was measured according to the
method described in14. The specimens were allowed to harden in molds for 48 h and then for 5 d in water. After this the
specimens were brought to harden in a desiccator at t= 20 ± 2 oC and R.H. = 65 %.
To determine water resistance, two sets of the AASC mortars were tested at the age of 28 d. The first set was allowed to
harden in normal conditions (t= 20 ± 2 oC, R.H. = 95 ± 5 %) and then compressive strength was determined. Another set
was preliminarily saturated with water. Coefficient of water resistance was determined as a ratio between average
strength of the specimens of two sets.
Compressive strength of the AASCs was determined as per EN 196-1 with taking into account the requirement of the
national standard of Ukraine DSTU B. V.2.7-181 in its part concerning choice of the ratio between sodium silicate
hydrate and slag in the mortars. This ratio was defined by flow values between 125 and 150 mm. Strength of the concrete
based on AASC was tested as prescribed by the national standard of Ukraine DSTU B.V.2.7-214.
The compositions of the AASCs are shown in Table 1. The modifiers (glycerin and a mixture of glycerin and hydrated
lime) were added directly into sodium silicate hydrate before mixing it with the slag. The total content of the liquid phase
(sodium silicate hydrate with modifiers) for all AASC compositions was assumed to be constant and equal to that of the
reference composition without modifiers (composition # 1).
Table 1. Cement compositions
Composition
#
Components
Slag
(% by mass)
Sodium silicate hydrate
modified by Na3PO4·12H2O
Glycerin
Hydrated lime
over 100 % of slag by mass
1 (reference)
100
38 – 55
-
-
2
100
5
-
3
100
5
1
4
100
10
-
5
100
10
1
2nd International RILEM/COST Conference on Early Age Cracking and
Serviceability in Cement-based Materials and Structures - EAC2
12–14 September 2017, ULB-VUB, Brussels, Belgium
3. RESULTS AND DISCUSSION
The calorimetric measurements of the reference AASC paste (Figure 1, curve # 1) and modified AASC pastes (Figure 1,
curves # 2 – # 5) allowed to show a correlation between early age structure formation in the AASC pastes and properties
of the AASC mortars (Table 2).
Table 2. Properties of the AASC pastes and mortars
# of
compositi
on
(see
Table 1)
Modifier
(% by mass)
Setting
time
(min)
Consistency
of mortar
(mm)
Compressive /flexural strength
(MPa)
Coefficient of water
resistance
Glycerin
Hydrated
lime
inital
final
3 h
1 d
7 d
28 d
1
-
-
15
17
126
21.5
4.6
32.3
5.0
66.5
7.2
86.8
9.0
0.88
2
5
-
30
40
135
15.9
2.8
28.2
4.7
65.6
7.7
88.0
10.5
0.92
3
5
1
30
50
138
15.8
3.3
30.2
5.7
64.8
8.5
95.5
11.7
0.95
4
10
-
50
80
142
7.5
2.2
18.0
3.6
55.5
7.4
83.8
10.0
0.90
5
10
1
45
55
140
9.5
2.5
27.4
4.4
60.5
8.4
89.8
10.9
0.93
A conclusion was drawn that in case of the reference AASC paste the induction period was almost absent after the initial
peak of heat evolution typical for wetting of the dispersed phase (i.e. slag). This phenomenon is attributed to the entropic
process during the AASC hydration and the formation of predominantly polymeric xerogel structures at the initial stage
of structure formation 15,16. This process precedes the formation of calcium silicate hydrates and affects negatively the
AASC hydration. As a result, the formation of a heterogeneous structure of the AASC stone with “fluctuating” values of
strength, deformation characteristics and coefficient of water resistance could be observed.
The necessity in physico-chemical impacts in order to obtain the AASC stone without decline of strength has been
reported in 15,16. At the initial stage of structure formation, silicic acid must be in a dissociated state to interact with
calcium ions, and polymerization of silicon-oxygen anions should occur later, i.e. during the period of agglomeration of
calcium silicate hydrates.
A dissociated state of the silicic acid during hardening of the AASC can be regulated by addition of glycerin or glycerin
with hydrated lime. Such approach makes it possible to control early age structure formation in the AASC pastes (Figure
1). When glycerin is added to the AASC (composition # 2) a noticeable induction period is fixed before setting (Table 2).
The calorimetric curves indicate conditions for the formation of primary nuclei of silicate phases and beginning of their
condensation (enthalpy process) with heat release (second peak). In this case, these processes precede polycondensation
of the xerogel structures.
The addition of hydrated lime in combination with glycerine (composition # 3) causes the lower total hydration heat, the
shorter induction period and shifts the second peak to the later age. This can be explained by topochemical interaction of
hydrated lime and sodium silicate hydrate, as well as by superposition of the formation of the silicate phases and
polycondensation of the xerogel structures.
2nd International RILEM/COST Conference on Early Age Cracking and
Serviceability in Cement-based Materials and Structures - EAC2
12–14 September 2017, ULB-VUB, Brussels, Belgium
a)
Q [J/g]
b)
dQ/dt [W/g]
Figure 1. Total heat (a) and rate of heat evolution (b) of AASC vs time.
The increased content of glycerin (composition # 4) and additional introduction of the hydrated lime (composition # 5)
was found to contribute to the longer induction period, retarded setting and the lower total hydration heat of the AASCs.
Thermokinetic processes during hydration of the AASCs correlate well with properties of the AASC mortars (Table 2).
Thus, the modifiers somewhat reduce the values of early strength (3 h) of the AASC stone, but after 7 days the strength is
sharply growing up and by 28 days the strength is close or higher than that of the reference composition. It is very
important that flexural strength of the mortars based on the modified AASCs in all cases exceeded the values of the
reference AASCs. Due to modification, the AASC mortars (concretes) obtain the higher water resistance.
The study of the AASC stone microstructure supports the above results and shows advantage of the modified AASC
against the reference AASC composition (Figure 2). A dense glass-like mass with separate inclusions of microcrystalline
secondary phases of silicate hydrates was formed in the structure of the reference AASC. The modified AASCs had
hydrated phases with more developed surface, which can be attributed to the formation of silicates and calcium
aluminosilicates.
Such microstructure affects positively crack resistance of the fine aggregate concrete and the shrinkage values are lower
(Figure 3). On example of the compositions # 1, # 2 and # 3, it can be clearly seen that addition of the proposed modifiers
allowed to reduce deformations by two and more times. The higher ratios between flexural and compressive strength to
values higher than 0.12, i.e. by 15 % higher than those of the reference AASCs, showed the higher crack resistance of the
modified AASC mortars. This showed a possibility to regulate the initial structure formation in the AASCs by
modification for reducing total hydration heat, adjustment of the induction period and providing the occurrence of the
second peak of the rate evolution curve.
2nd International RILEM/COST Conference on Early Age Cracking and
Serviceability in Cement-based Materials and Structures - EAC2
12–14 September 2017, ULB-VUB, Brussels, Belgium
Composition # 1
Composition # 2
Composition # 3
Figure 2. Microstructure of the AASC stone at 28 days.
Figure 3. Drying shrinkage of the fine aggregate AASC concretes.
4. CONCLUSION
The results of the study suggested to show that calorimetry could be a reliable method for prediction of early age
structure formation in the AASC pastes which affects crack resistance of the early high strength concretes based on
alkali-activated slag cements in the system "granulated blast-furnace slag - modified high modulus sodium silicate
-1.2
-1
-0.8
-0.6
-0.4
-0.2
0
8 d
(«0»)
14 d 28 d 35 d 42 d 49 d 56 d 63 d 70 d 78 d 86 d 93 d 100 d
Age [d]
Shrinkage [mm/m]
# 1
# 2
# 3
2nd International RILEM/COST Conference on Early Age Cracking and
Serviceability in Cement-based Materials and Structures - EAC2
12–14 September 2017, ULB-VUB, Brussels, Belgium
hydrate". A conclusion was drawn that chosen modifying admixtures affected positively the composition and
morphology of the hydration products. This modification allowed to produce a durable structure of the fine aggregate
concretes based on the alkali-activated slag cements with the following performance properties of the resulted concrete:
shrinkage 0.6 mm/m, coefficient of water resistance 0.88, compressive strength 15 MPa at age of 3 h and 80 MPa
at age of 28 days. Effect of each chosen modifying admixture could be evaluated by total hydration heat of the alkali-
activated slag cements and duration of the induction period before the occurrence of the second peak of rate evolution
curve.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the contribution of the COST Action TU 1404 “Towards the next generation of
standards for service life of cement-based materials and structures” (http://www.tu1404.eu). The authors also express
their gratitude to the Ministry of Education and Science of Ukraine for financial support of this research that was carried
out under the ranges of the following topics No. 0115U000715 and No. 0116U000842.
REFERENCES
1. Shi C., Krivenko P., Roy D.: Alkali activated cements and concretes, Taylor&Francis, Abingdon; 2006.
2. Palomo A., Krivenko P., Garcia-Lodeiro I., Kavalerova E., Maltseva O., Fernández-Jiménez A. A review on alkaline
activation: new analytical perspectives, Materiales de Construcción. Vol. 64 (315), 1-24 (2014)
3. Krivenko P., Runova R., Sanуtsky M., Rudenko I. Alkaline cements. Monograph, Kyiv, Publishing house "Osnova";
2015.
4. Krivenko P.: Synthesis of cementitious materials of the Me2O-MeO-Me2O3-SiO2-H2O system with required
properties, DSc (Eng) Thesis, Kyiv; 1986.
5. Solodkyy S.: Alkali-activated slag cement concrete for the construction of pavements. PhD Thesis: Kyiv; 1991.
6. Pacheco-Torgal F., Barroso de Aguirre J., Ding Y., Tahri W., Baklouti S.: Performance of alkali activated mortars
for the repair and strengthening of OPC concretes. Handbook of Alkali-activated Cements, Mortars and Concretes”
Edited by Pacheco-Torgal et al, Elsevier. 627- 642 (2015)
7. Krivenko P., Petropavlovsky O., Petranek V., Pushkar V., Vozniuk G.: High strength alkali activated slag cements
with controlled setting times and early strength gain. Advanced Materials Research. Vol. 1100, 44-49 (2015)
8. Krivenko P., Petropavlovskii O., Pushkar V.: Methods of regulation of the properties of alkali slag cements and
concretes based on the liquid-glass. Proceedings of 18th Ibausil Internationale Baustofftagung. Weimar,
Tagungsbericht. Band 2.– S. 1-1178-1-1185 (2012)
9. http://эволют.рф/pages/view/56 (last accessed 2012)
10. Korneev V., Danilov V.: Water and Soluble Glass. Stroyizdat, Saint-Petersburg; 1996.
11. Krivenko P., Petropavlovsky O., Gelevera A., Vozniuk G., Lakusta S.: The influence of organic-mineral modifiers
on structure formation and properties of high strength alkali- activated slag cements with soluble silicates. Building
materials and products. No. 4, 38-43 (2016)
12. Usherov-Marshak A.: Calorimetry of cement and concrete, Kharkiv: Fakt; 2002.
13. Usherov-Marshak A., Kabus A.: Calorimetric monitoring of early hardening of cement in the presence of
admixtures. Inorganic Materials. Vol. 49, No. 4, 449-452 (2013)
14. Butt Y., Timashev V.: Workshop on chemical technology of binders, Moskow: Vysshaya shkola; 1973.
15. Yaglov V., Burak G., Mezhentsev A.: Nanoparticles in concrete. Building science and technology. No. 1, 21-30
(2012)
16. Vyrodov I.: Thermodynamics and thermochemistry of cement. Proceedings of 4th International Congress on the
Chemistry of Cement. Moscow: Stroyizdat. 40-41 (1976)











