Composites: Part B 57 (2014) 166–172

Contents lists available at ScienceDirect

Composites: Part B
journal homepage: www.elsevier.com/locate/compositesb

Effect of activator type and content on properties of alkali-activated

slag mortars
Serdar Aydın ⇑, Bülent Baradan
Department of Civil Engineering, Dokuz Eylül University, Buca 35160, Izmir, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: Investigation of alkali activation of Turkish slag (AAS) was carried out using sodium silicate and sodium
Received 7 October 2012 hydroxide activators within the scope of this study. The objective of the present work is to determine the
Received in revised form 17 June 2013 SiO2/Na2O ratios (MS) and Na2O contents of the solutions on the development of workability, setting
Accepted 1 October 2013
times, mechanical properties, drying shrinkage, water absorption characteristics and microstructure of
Available online 10 October 2013
alkali activated slag cement binders. Test results showed that MS and Na2O contents of activator solution
are of great importance on the properties of AAS. Portland cement free high performance composite with
Keywords:
compressive strength values about 100 MPa can easily be achieved by activation of slag without heat cur-
B. Mechanical properties
B. Microstructures
ing. Moreover, in case of activation by optimum MS ratio, sodium silicate activated AAS mortars present
Alkali-activated cement higher compressive strength, lower water absorption, higher workability, lower porosity and a wide
range of setting times compared to NaOH activated AAS mortars and Portland cement mortar. Conse-
quently, it can be said that this new binder is likely to have enormous potential to become an alternative
to Portland cement.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction more, questionable results exist about alkali–aggregate reactions

and carbonation [2–4].
Ground granulated blast furnace slag (GGBFS) is a by-product of Since rather early discovery of alkali-activated cements and
steel production. It has been reported that, slag generation is about concretes in 1958, they have been commercially produced and
10% of the total production. Blast furnace slags are usually utilised used in limited construction projects in the former Soviet Union,
in construction industry in the production of blended cements and China and some other countries. AAS have been used in the con-
concretes. However, only small amounts (20–30%) of slags are uti- struction of drainage anti-soil slipping collector/tubing, silage
lised in this area and the rest is stored in large volumes [1]. The trenches, slopes of railway embankment, high storey residential
production of high performance alkali activated slag (AAS) based buildings, special concrete pavements for heavy loaded trucks, pre-
composites may be an alternative utilisation area of this by- stressed reinforced concrete slippers for railways, masonry blocks,
product. oil-well concreting, refractory concrete, stabilization and solidifica-
Utilisation of GGBFS in this area has significant environmental tion of hazardous and radioactive wastes, etc. Various inspections
and economical benefits since its production requires less energy during 1999–2000 years showed that in all these structures, the
than the production of Portland cement. And there is no CO2 emis- performance of alkali-activated slag cement concrete was superior
sion to the atmosphere during this process. Besides these environ- compared to the Portland cement concretes used in the same area
mental and economical benefits, AAS binders present some [5]. However, AAS is not a well-known binder in Turkey besides a
important technical advantages over ordinary Portland cements. very limited research.
These can be listed as; the development of earlier and higher Superior mechanical properties have been reported for AAS
mechanical strengths, lower heat of hydration, low porosity, low binders. In China, compressive strength of 60–150 MPa for AAS
solubility of the hydrates, stronger aggregate–matrix interface, concrete activated by waterglass was achieved without heat treat-
and superior durability. However, they also have some disadvan- ment or special additives [6]. Zivica et al. reported a compressive
tages such as rapid setting, high shrinkage, subsequent formation strength of 150 MPa for alkali-activated slag paste prepared with
of microcracks and higher formation of salt efflorescences. Further- alkali activator solution: slag ratio of 0.08 and pressure compaction
in fresh state [7].
Activation process of slag varies with the chemical properties
and phase compositions of slags, the type and concentration of
⇑ Corresponding author. Tel.: +90 232 301 7044; fax: +90 232 301 7253.
activators. Slags have a variable composition depending on the
E-mail address: serdar.aydin@deu.edu.tr (S. Aydın).

1359-8368/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.compositesb.2013.10.001
 S. Aydın, B. Baradan / Composites: Part B 57 (2014) 166–172 167

raw materials and the industrial process; hence, each slag response were demoulded 24 h after casting and cured in water for 48 h.
differently to the activation process. For every slag it is necessary, The first reading of length was taken at the age of 72 h. After the
to plan a research program for determining the most suitable acti- first reading, the prisms were placed in laboratory (20 °C temper-
vator type and content. Sodium hydroxide and sodium silicate ature and 55% relative humidity). Length changes of prisms were
(water glass) are the most commonly used activators. Sodium sili- measured periodically up to 6 months. Capillary water absorption
cate is also considered the most effective activator. The most of mortars was determined according to ASTM C1403 [13]. Water
important properties of water glass are the ratio of SiO2/Na2O absorption and volume of permeable voids of mortars were also
and Na2O content. The aim of this study is to determine the effect determined according to ASTM C642 [14]. Initial and final setting
of these parameters on the activation of a Turkish slag in respect to times were determined by using a Vicat apparatus according to
the mechanical properties, microstructure, workability, setting TS EN 196-3 [15]. The water/binder ratio of all the pastes was kept
times, water absorption and shrinkage characteristics. constant at 0.33.

2. Materials and experimentation 3. Results

Ground granulated blast furnace slag procured from Ereğli The effectiveness of MS and Na2O content of solution on the
steel plant has a chemical composition presented in Table 1. activation of GGBFS was investigated by various tests on different
Specific gravity and specific surface (Blaine) of GGBFS is 2.88 mortar mixtures. Test results are discussed below:
and 410 m2/kg, respectively. It contains, 90% particles sized
lower than 45 lm. It is neutral with the basicity coefficient 3.1. Workability
[Kb = (CaO + MgO)/(SiO2 + Al2O3)] of 0.81. As can be seen from
Table 1, the hydration modulus HM = (CaO + MgO + Al2O3)/SiO2 of Flow values of fresh mortars according to their MS and Na2O val-
slag is 1.33. ues are given in Fig. 1 together with control PC mortars. PC mortars
Ordinary Portland cement (CEM I 42.5R) with a Blaine fineness are marked by dash line in all figures. As can be observed from
of 369 m2/kg has been used as reference binder. Chemical compo- Fig. 1, most of the AAS mortars present higher workability than
sition of Portland cement (PC) has also been presented in Table 1. PC mortars. Also, flow values of mortars increased parallel to the
Alkali activated slag (AAS) mortars were produced by activation increase in MS at relatively low Na2O contents (2% and 4%).
of GGBFS with technical grade sodium hydroxide and water glass However, at higher Na2O contents this trend was not observed
(liquid sodium silicate). Sodium silicate used in this study has a especially in MS = 0.4 and 0.8. For MS = 0 case, higher Na2O contents
27% and 8% of SiO2 and Na2O contents, respectively and with a resulted in better flow values.
MS value (SiO2/Na2O) of 3.38. Sodium hydroxide and sodium sili-
cate were mixed in various proportions, within the MS range of
3.2. Setting times
0.4–1.6. Also, solutions with MS equal to zero have been prepared
by using NaOH alone. AAS mixtures with Na2O contents of 2%,
Initial and final setting times of pastes for various MS and Na2O
4%, 6%, and 8% were investigated after pre trials. Activator solutions
values are presented in Fig. 2. Setting times of AAS pastes varied in
were prepared 1 day before casting of the mortars.
a wide range. Initial setting times were recorded in the range of 2–
Twenty different alkali-activated mortar mixtures and a control
326 min, while final setting times were 2–442 min. It should be
PC mortar mixture were prepared with crushed limestone sand (0–
stated that, standard consistencies of (6% Na2O, MS = 0.4 paste)
4 mm). Aggregate to binder (PC or GGBFS) ratio of 2.75 was kept
and (8% Na2O, MS = 0.4 and 0.8 pastes) were between 20 and
constant for all mixtures. Water to binder ratio of all mixture
31 mm. The standard consistencies of the other mixtures were
was 0.44. All test batches were mixed by using an electrically dri-
lower than 7 mm, similar to PC paste’s consistency.
ven mechanical mixer conforming to the requirements of ASTM
For 4% Na2O mixture, higher MS ratios caused a slight reduction
C305 [8]. Initially, binder and aggregate were mixed in a dry state
in the setting times (Fig. 2). For mixtures with 6% and 8% Na2O con-
for 1 min and then the solutions were gradually added during mix-
tents, setting times of pastes decreased dramatically for MS = 0.4
ing for about 3 min. The flow test was performed according to
and 0.8. These mixtures seem to be not appropriate for normal con-
ASTM C230 [9]. Fresh mixtures were cast into prismatic
struction works. However, they may be very appropriate for the
(40  40  160 mm and 25  25  285 mm) steel moulds. The
works that needs quick setting, such as shotcrete. The faster setting
specimens were kept in the moulds in the humidity cabinet for
of AAC by sodium silicate may be attributed to the formation of an
24 h at about 20 °C and a relative humidity of 90%. After demoul-
ding, specimens were kept in water at a temperature of 20 °C until
the testing periods. 160
The prismatic specimens (40  40  160 mm) were subjected
to flexural strength test according to ASTM C348 [10]. Three 150
specimens were tested and average values were recorded. The
specimens were loaded from their mid-span and the clear distance
Flow value (mm)

140
between simple supports was 120 mm. The compressive strength
tests were performed according to ASTM C349 [11]. Shrinkage 130
measurements were performed on mortar bars
(25  25  285 mm) according to ASTM C596 [12]. Specimens 120

110
Table 1
Chemical composition (%) of GGBFS and cement. 2% Na2O 4% Na2O 6% Na2O 8% Na2O PC
100
SiO2 Fe2O3 Al2O3 CaO MgO Na2O K2O SO3 LOI 0 0.4 0.8 1.2 1.6

GGBFS 40.20 1.68 11.66 35.90 5.88 0.30 1.47 0.90 0.88 Ms
PC 19.10 3.96 4.40 61.85 2.05 0.27 0.70 3.72 1.82
Fig. 1. Flow values of mortars.
168 S. Aydın, B. Baradan / Composites: Part B 57 (2014) 166–172

Fig. 2. Initial and final setting times of pastes.

initial calcium silicate hydrate. Zivica [16] has also stated that Table 3
setting in pastes activated by sodium hydroxide is the result of Mechanical properties of PC mortars.

formation of extra polysilicates hydrates. The setting times of the Time (day) Compressive strength (MPa) Flexural strength (MPa)
pastes were prolonged by increasing MS ratios to 1.2 and 1.6. As 2 39.7 9.8
a general conclusion it may be stated that; except for intermediate 7 47.7 9.9
MS values, higher Na2O contents results in longer setting times. 28 55.1 11.8
90 61.7 13.1

3.3. Mechanical properties

decrease of C/S ratio in C–S–H structure and causes an improve-
Compressive and flexural strength values of AAS and PC mortars
ment in binding ability of C–S–H.
are given in Tables 2 and 3, respectively. As can be observed from
Tables 2 and 3, the mechanical properties of AAS mortars are gen-
erally higher than conventional PC mortars at later ages. 2% Na2O 3.4. Drying shrinkage
incorporated AAS mortars disintegrated in water; hence mechani-
cal properties of these mortars could not be determined. Also, 4% Drying shrinkage of 4%, 6% and 8% Na2O incorporated AAS mor-
Na2O incorporated mortars with high MS ratios did not gain any tars were observed up to 6 months. Drying shrinkage behaviour of
mechanical strength at early ages (2-day). 6% Na2O incorporated AAS mortars and PC mortar versus time are
Compressive strength values increased with the increasing MS given in Fig. 3. As shown in Fig. 3, drying shrinkage of these mor-
ratios, however flexural strength did not changed remarkably. tars remained approximately at the same level beyond 40 days.
The optimum MS values for flexural and compressive strengths Similar behaviour was also observed for 4% and 8% Na2O incorpo-
differ both for early and later ages. Higher early strengths were rated AAS mortars. Drying shrinkage of all AAS mortars at
obtained for lower MS values in contrast to later ages. For example, 6 months are presented in Fig. 4. As can be observed from Figs. 3
at 6% Na2O concentration, maximum compressive and flexural and 4, mortars that have MS value of 0 showed similar drying
strength values at 2 days were obtained for MS values of 0.4 and shrinkage behaviour with PC mortars. Higher ratios of MS resulted
0, respectively. And at later ages (90-day), maximum values in higher drying shrinkage values significantly. Nevertheless, none
recorded for MS values were 1.6 and 1.2, respectively. This may of the specimens exhibit visible shrinkage cracks by naked eyes.
be due to the incorporation of limited amount of silicate anions Na2O content seems to have no effect on drying shrinkage of mor-
to C–S–H structure at early ages. Parallel to the hydration process, tars up to MS value of 0.4. However, above this value, drying
silicate anions ingress into the C–S–H structure. This results in a shrinkage also increased with the increasing Na2O content. As a

Table 2
Mechanical properties of AAS mortars.

MS 0 0.4 0.8 1.2 1.6 0 0.4 0.8 1.2 1.6

Na2O, % Compressive strength (MPa) Flexural strength (MPa)
2-day
4 8.2 14.3 6.5 0.0 0.0 4.0 2.5 1.3 0.0 0.0
6 11.6 22.6 17.3 14.5 12.5 6.6 6.0 4.3 2.4 1.6
8 10.9 11.3 10.1 18.9 17.7 5.6 2.2 2.2 2.9 2.9
7-day
4 14.3 38.4 35.5 8.6 4.3 6.8 7.8 5.2 2.8 1.1
6 17.8 46.1 40.7 40.2 37.2 9.2 9.0 6.6 4.3 4.2
8 18.5 34.7 43.9 52.6 44.9 8.4 6.8 5.7 4.8 4.4
28-day
4 21.3 54.7 59.7 58.5 52.3 8.6 11.3 12.5 7.9 8.5
6 24.6 62.4 78.4 74.1 71.0 10.1 11.4 11.3 10.8 9.9
8 25.3 48.3 70.3 81.7 74.1 11.0 8.3 8.2 10.0 8.5
90-day
4 26.2 61.3 84.0 67.8 57.3 7.8 10.5 12.5 11.9 9.9
6 31.6 75.2 76.1 87.2 88.3 8.8 11.8 12.7 13.4 10.5
8 37.3 60.6 63.0 94.8 99.0 11.1 9.9 12.1 11.6 10.6
 S. Aydın, B. Baradan / Composites: Part B 57 (2014) 166–172 169

6000 to 8%, resulted in an increase of capillary water absorption. Low

Ms=0 Ms=0.4 Ms=0.8 Ms=1.2 Ms=1.6 PC
workability and quick setting properties of 8% Na2O incorporated
5000 mortars for MS values of 0.4 and 0.8 may be effective on this behav-
Drying shrinkage (x10 )
-6

iour. These fresh mortar mixtures were hard to place and compact
4000
in the moulds. A more porous structure may form due to this
3000
reason.
Total water absorption values (TWA) and permeable voids (PV)
2000 of AAS mortars are presented in Fig. 6. Contrary to capillary water
absorption case, some AAS mortars present lower water absorption
1000 values and have less permeable voids compare to PC mortars. It is
believed that this phenomenon is related to the pore size distribu-
0 tion of mortars. It is known that; higher pore radii results in lower
0 20 40 60 80 100 120 140 160 180
Time, days
coefficient of capillary water absorption values. This indicates that
AAS mortars have smaller pore radii compared to PC mortars. This
Fig. 3. Drying shrinkage of 6% Na2O incorporated mortars and PC mortar. phenomenon has also been confirmed by microstructure investiga-
tions. Similar to capillary water absorption case, water absorption
and volume of permeable voids values were minimum at MS = 0.8.
6000
4% Na2O 6% Na2O 8% Na2O
Drying shrinkage at 6 months (x10 )

3.6. Microstructure investigations

-6

5000

4000
Microstructure investigations on mortars were carried out by
X-ray analyses, SEM investigations and mercury intrusion porosimetry
(MIP). These investigations were performed on 6% Na2O incorpo-
3000
rated mortar samples with MS = 0, 0.4 and 1.2 at 28-day age.
SEM investigations were applied on fractured surfaces by sec-
2000
ondary mode (SE) and polished sections by backscattered mode
(BSE). Fig. 7 presents the SEM (SE) images of AAS mortars. As
1000
shown in Fig. 7, the hardened paste of NaOH and water glass acti-
vated mortars exhibit quite different microstructure under SEM. In
0
0 0.4 0.8 1.2 1.6
case of NaOH activation (Fig. 7a), the microstructure is highly por-
Ms
ous with the increasing MS ratios the structure becomes well
packed and massive (Fig. 7b and c). All mortars display microcracks
Fig. 4. Drying shrinkage of AAS mortars at 6 months. in the interfacial aggregate–paste zone. However, cracks intensity
increases with the higher values of MS, both in the matrix and in
the aggregate–matrix transition zone. This microstructural analy-
conclusion, it can be said that; MS value is more effective than Na2O ses result is valid for all the pastes, the main reaction product is
content on drying shrinkage properties of AAS mortars. a calcium silicate hydrate (with the decreased Ca/Si from 2.88 to
1.38 by increasing MS ratio). The EDS analyses also revealed that
3.5. Capillary water absorption, total water absorption and permeable C–S–H structure consists Na element. The amount of Na incorpo-
rated into C–S–H structure decreased from 7.3% to 3.9% by increas-
voids
ing MS from 0 to 1.2. Some calcite crystals have also been detected
in the case of MS = 1.2 (Fig. 7c).
Capillary water absorption values of AAS mortars at 24 h are
SEM (BSE) images of AAS are given in Fig. 8. Light grey areas
presented in Fig. 5. As can be seen in Fig. 5, capillary water absorp-
shows unhydrated slag grains, dark grey areas represent limestone
tion of all AAS mortars is higher than PC mortars. Capillary water
aggregates, and black areas show the existence of pores and cracks.
absorption of AAS mortars generally showed a minima at
Images indicate existence of many unreacted slag particles. AAS
MS = 0.8. Increment of Na2O content from 4% to 6% decreased the
mortar activated by NaOH presents a more porous formation
capillary water absorption. However, increasing of Na2O content
in contrary to mortars activated by water glass. Although AAS

120
12 20
2
Capillary water absorption, g/100 cm

100 11 18
Volume of permeable voids, %...
Total water absorption, %

10 16
80
9 14

60 8 12

7 10
40
6 8
20
5 6
4% Na2O (TWA) 6% Na2O (TWA) 8% Na2O (TWA) PC (TWA)
4% Na2O 6% Na2O 8% Na2O PC 4% Na2O (PV) 6% Na2O (PV) 8% Na2O (PV) PC (PV)
0 4 4
0 0.4 0.8 1.2 1.6 0 0.4 0.8 1.2 1.6
Ms Ms

Fig. 5. Capillary water absorption (at 24 h) of mortars. Fig. 6. Water absorption and volume of permeable voids.
170 S. Aydın, B. Baradan / Composites: Part B 57 (2014) 166–172

(a) Ms=0
Aggregate

Matrix

(b) Ms=0.4
Aggregate

Matrix

(c) Ms=1.2
Aggregate

Calcite
crystals
Matrix

Fig. 7. SEM (SE) images of mortars incorporated 6% Na2O (left sides matrix, right sides aggregate–matrix interface).

mortars activated by water glass have well-packed and massive not meet to Japanese standard, relatively high compressive
structure, they also show a higher ratio of microcracks due to the strength values about 100 MPa can be obtained by the alkali acti-
tensions caused by excessive shrinkage. The cracks pass through vation of this type slag.
the matrix of reaction products and through the faces of the The main reaction product of AAS is Na incorporated C–S–H
unreacted slag particles (see Fig. 8b and c). with low Ca/Si ratio. Malolepszy [18] postulated the formation of
Another crystal apart from calcite was not detected in AAS mor- a solid solution of Na2O–CaO–SiO2–H2O (NCSH), since Na+ ions in
tars by XRD analyses. The origin of calcite is the limestone aggre- alkali-activated cement have a very low solubility in water.
gate. In other words, AAS mortars exhibit an amorphous Compressive strength of AAS mortars increased significantly
structure in XRD analyses. with the increasing values of MS. The main reasons for this result
As shown in Fig. 9, MS = 0 mortar has a coarser pore size distri- can be summarised as:
bution, and has considerably more macropores and mesopores
compared to AAS mortars activated by water glass. For water glass – Porous and poorly packed structure transforms into a dense and
activated mortars, the amount of mesopores for MS = 0.4 mortar is compact structure by increasing MS value from 0 to 0.4 and 1.2.
about one half of the MS = 1.2 mortar. The amount of macropores – Ca/Si ratio of C–S–H decreased with the higher values of MS. The
for MS = 0.4 mortar is also lower than MS = 1.2 mortar. binding ability of C–S–H improves with a decrease in the Ca/Si
ratio [19].
– MIP measurements showed that porosity decreases with higher
4. Discussions MS values.

Hydration modulus (HM) of slag is a quality criterion for blast It is well known that, flexural strength is more sensitive to
furnace slag for many countries. For example, HM should be great- cracks than compressive strength. Distinct from the case of com-
er than 1.0 in European and Canadian standards, and should be pressive strength, flexural strength did not change remarkably
greater than 1.4 in Japanese standard [5]. It is also suggested that with the increasing values of MS. SEM images showed a significant
the HM of slag should exceed 1.4 to ensure sound hydration increase of microcracks in the matrix and matrix–aggregate inter-
property [17]. Although slag used in this study (HM = 1.33) does faces in case of high MS values.
 S. Aydın, B. Baradan / Composites: Part B 57 (2014) 166–172 171

(a) Ms=0 Na2O concentration in contrast to later ages. This finding was also
reported by Bakharev et al. [6].
Alkali-activated slag concrete exhibits higher drying shrinkage
Aggregate
than ordinary Portland cement concrete [20–22]. Wittmann [23]
and Young [24] show that the pore size distribution and the cal-
cium silicate hydrate gel characteristics have a critical influence
Unhydrated
GGBFS on the magnitude of drying shrinkage. Higher drying shrinkage of
water glass activated mortars (MS > 0) compare to NaOH activated
mortars (MS = 0) may be attributed to the differences in micro-
structure detected in SEM (Fig. 8). Shrinkage under various condi-
tions depends on the loss of water from the mesopores (pore
diameter between 2.5 and 50 nm) and also the size of the macrop-
ores (pore diameter between 50 and 10,000 nm), which deter-
mines easiness of evaporation of water from the mesopores [21].
For the water glass activated mortars, higher drying shrinkage val-
(b) Ms=0.4 ues of MS = 1.2 mortar compared to MS = 0.4 may be attributed to
existence of higher volume of mesopores.
For reasonable working periods, initial setting times should not
be less than 30 min and final setting times should not exceed 10 h
for different types of Portland Cements. A quick-setting cement
must have an initial setting time less than 5 min and a final setting
time of not more than 30 min [6]. According to these criteria, 6%
and 8% Na2O incorporated pastes with MS = 0.4 and 0.8 exhibit set-
ting times are similar to quick-setting cements whereas the others
complies with the setting time limits for normal setting Portland
cements.

5. Conclusions

(c) Ms=1.2 Based on the results of this experimental investigation, the fol-
lowing conclusions are drawn:

– Alkali concentration and silicate modulus of an activator have a

vital importance on the setting times, workability, mechanical
properties and microstructure of AAS based binders. AAS mor-
tars properties can be arranged by changing of the activator
type and its content.
– In case of optimum MS values, AAS mortars activated by water
glass presents higher compressive strength, lower water
absorption characteristics, higher workability, lower porosity
and a wide range setting times compare to AAS activation with
NaOH and PC mortar. On the other hand, drying shrinkage of
these mortars is considerably higher than NaOH activated AAS
and PC mortars. Higher alkali contents and especially higher sil-
Fig. 8. SEM (BSE) images of mortars incorporated 6% Na2O.
icate modulus caused higher drying shrinkage values for these
mixtures.
– At high alkali concentrations (6% and 8% Na2O) for intermediate
0.08
MS values (0.4 and 0.8), workability and setting times of AAS
Cumulative volume intruded (cm 3)

0.07 Macropores Mesopores mixtures are not suitable for normal construction works.
0.06 – Mechanical properties of AAS mortars generally increased
MS=0
with the higher alkali contents. Higher early strengths were
0.05
obtained in case of low modulus solutions in contrast to at
0.04 later ages.
0.03 – Early age mechanical properties of AAS mortars were lower
0.02 MS=1.2
than PC mortars remarkably. However, at later ages compres-
sive and flexural strength of AAS mortars exceeded the
0.01
MS=0.4
strength of PC mortars. A comparatively high compressive
0 strength value of 100 MPa was found for the AAS mortar acti-
100000 10000 1000 100 10 1
vated by water glass without heat treatment or special addi-
Pore diameter (nm)
tives. This result show that properly designed AAS mortars
Fig. 9. Pore size distribution of AAS mortars. may be a promising binder for the production of high perfor-
mance composites.
– Minimum values of capillary water absorption, total water
Higher early strengths both compressive and flexural were ob- absorption and volume of permeable voids of AAS mortars has
tained for the solutions with low modulus for all three levels of been found for the mixture MS = 0.8 at 6% Na2O content.
172 S. Aydın, B. Baradan / Composites: Part B 57 (2014) 166–172

– Reaction products of AAS exhibit an amorphous structure in [8] ASTM C 305–99. Standard Practice for Mechanical Mixing of Hydraulic Cement
Pastes and Mortars of Plastic Consistency. Annual Book of ASTM, USA; 2002.
XRD. The main reaction product of AAS independent from MS
[9] ASTM C 230/C230M-98. Standard Specification for Flow Table for Use in Tests
is Na incorporated C–S–H. However, Na content and Ca/Si ratio of Hydraulic Cement. Annual Book of ASTM, USA; 2002.
of this C–S–H structure decreases by increasing of MS. [10] ASTM C 348–97. Standard Test Method for Flexural Strength of Hydraulic-
– Raise in MS caused to formations of microcrack in the matrix Cement Mortars. Annual Book of ASTM, USA; 2002.
[11] ASTM C 349–97. Standard Test Method for Compressive Strength of Hydraulic-
and the matrix–aggregate interface. Compressive strength val- Cement Mortars (Using portions of prisms broken in flexure). Annual Book of
ues increased parallel to the raise in MS ratios, whereas flexural ASTM, USA; 2002.
strength values remained almost constant. This phenomenon [12] ASTM C 596–01. Standard Test Method for Drying Shrinkage of Mortar
Containing Hydraulic Cement. Annual book of ASTM standards, USA; 2004.
may be explained by the extra crack sensibility of flexural stress [13] ASTM C 1403–05. Standard Test Method for Rate of Water Absorption of
compared to compressive stress case. Masonry Mortars. Annual book of ASTM standards, USA; 2006.
[14] ASTM C 642–97. Standard Test Method for Density, Absorption, and Voids in
Hardened Concrete. Annual book of ASTM standards, USA; 2004.
[15] TS EN 196–3. Methods of Testing Cement Part 3: Determination of Setting
Acknowledgement Times and Soundess. Turkish Standards Institution, Turkey; 2002.
[16] Zivica V. Effects of type and dosage of alkaline activator and temperature on
the properties of alkali-activated slag mixtures. Construct Build Mater
This study is a part of the project supported by the Scientific Re- 2007;21:463–9.
search Council of Dokuz Eylul University (2009.KB.FEN009). The [17] Chang JJ. A study on the setting characteristics of sodium silicate-activated
authors gratefully acknowledge for their financial support. slag pastes. Cem Concr Res 2003;33:1005–11.
[18] Malolepszy J. Some aspects of alkali activated cementitious materials setting
and hardening. In: 3rd Beijing international symposium on cement and
References concrete, vol. 2, 1993. p. 1043-1046.
[19] Bakharev T, Sanjayan JG, Cheng YB. Effect of admixtures on properties of alkali-
[1] Puertas F, Martinez-Ramirez S, Alonso S, Vazquez T. Alkali-activated fly ash/ activated slag concrete. Cem Concr Res 2000;30:1367–74.
slag cement: strength behaviour and hydration products. Cem Concr Res [20] Collins FG, Sanjayan JG. Workability and mechanical properties of alkali
2000;30:1625–32. activated slag concrete. Cem Concr Res 1999;29:455–8.
[2] Fernandez-Jimenez A, Palomo JG, Puertas F. Alkali-activated slag mortars- [21] Collins FG, Sanjayan JG. Effect of pore size distribution on drying shrinkage of
mechanical strength behaviour. Cem Concr Res 1999;29:1313–21. alkali-activated slag concrete. Cem Concr Res 2000;30:1401–6.
[3] Bakharev T, Sanjayan JG, Cheng YB. Effect of elevated temperature curing on [22] Atısß CD, Bilim C, Çelik Ö, Karahan O. Influence of activator on the strength and
properties of alkali-activated slag concrete. Cem Concr Res 1999;29:1619–25. drying shrinkage of alkali-activated slag mortar. Construct Build Mater
[4] Zivica V. Effectiveness of new silica fume alkali activator. Cem Concr Compos 2009;23:548–55.
2006;28:21–5. [23] Wittmann FH. Creep and shrinkage mechanisms. In: Bazant ZP, Wittmann FH,
[5] Shi C, Krivenko PV, Roy D. Alkali-activated cements and concretes. USA and editors. Creep and shrinkage in concrete structures. Chichester: Wiley; 1982.
Canada: Taylor and Francis; 2006. p. 129–61.
[6] Bakharev T, Sanjayan JG, Cheng YB. Alkali activation of Australian slag cements. [24] Young JF. Physical mechanisms and their mathematical descriptions. In:
Cem Concr Res 1999;29:113–20. Bazant ZP, editor. Mathematical modelling of creep and shrinkage of
[7] Zivica V, Bagel L, Kuliffayova M. High strength alkali-activated slag paste. Build concrete. Chichester: Wiley; 1988. p. 63–98.
Res J 2010;58:175–86.