Construction and Building Materials 25 (2011) 1018–1025

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Microstructure and durability of mortars modified with medium active

blast furnace slag
Ahmed Hadj-sadok a,b, Said Kenai a,*, Luc Courard c, Anne Darimont c
a
Geomaterials Laboratory, Civil Engineering Department, University of Blida, Algeria
b
Hydraulics National Institute (ENSH), Blida, Algeria
c
GeMMe Sector, ArGEnCo Department, Faculty of Applied Science, University of Liège, Belgium

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

Article history: Mechanical characteristics and durability properties of blast furnace slag cement composites largely
Received 13 February 2010 depend on the hydraulic activity of the slag. In this paper, a Granulated Blast Furnace Slag with a low
Received in revised form 5 April 2010 reactivity index is used in modifying mortar composition. Microstructure and durability of mixes con-
Accepted 19 June 2010
taining 0%, 30% and 50% of slag as substitution to OPC are respectively compared and analyzed. Water
porosity, Mercury Intrusion Porosity and pore size distribution are studied after 28, 90 and 360 days of
wet curing. A qualitative microstructure analysis of mortars is proposed with Scanning Electron Micro-
Keywords:
scope (SEM). The durability of mortar is evaluated through capillary water absorption and chloride dif-
Environment
Mortar
fusion tests. The results indicate a finer porosity and lower water absorption for slag mortars at old
Blast furnace slag ages (90 and 360 days). Moreover, lower chloride diffusion for 50% blast furnace slag substitution is
Hydraulic activity observed.
Pore structure Ó 2010 Elsevier Ltd. All rights reserved.
Microstructure
Durability
Chloride diffusion

1. Introduction performance and durability of mortars and concretes are

The use of cementitious by product materials has become a
common tendency. It is an efficient procedure for the cement
industry in order to decrease CO2 emissions, which represents
about 5% of the total anthropic emissions in the world [1]. In addi-
tion, it can contribute to save natural resources, recycle by-prod-
ucts and preserve the environment.
Ground granulate blast furnace slag (GGBFS) is a by-product
from the iron industry; it is one of the most commonly used addi-
tions in the cement industry. According to the manufacturing pro-
cess, GGBFS is a quite variable material due to the variability of its
chemical composition. Offering latent hydraulic properties when
mixed with clinker cement, hydration of slag is directly related
to its hydraulicity: the dissolution of slag glass fraction is ensured
by hydroxyl ions (OH ) resulting from the hydrolysis of Portlandite
Ca(OH)2 produced by the hydration of clinker [2,3]. The hydration
products formed in the cement matrix are mainly additional hy-
drated calcium silicates and aluminates (CSH, CAH). The resulting
hydrated cementitious matrix presents good chemical resistance
and a more refined pores structure [4,5]. Therefore the mechanical
* Corresponding author. Tel./fax: +213 25433939.
E-mail address: sdkenai@yahoo.com (S. Kenai).
improved.
Guneyisi and Gesoglu [6] indeed observed a decrease of capil-
lary absorption for concrete containing high levels of slag (>50%).
Moreover, Shi et al. [7] also observed an improvement of the
behaviour of 50% slag concrete against gas permeability for pro-
longed wet curing. Finally, according to Yun Yeau et al. [8], the
coefficient of diffusion of a concrete with 55% slag was three times
smaller than that of OPC concrete.
Nevertheless, this improvement in durability properties de-
pends on the good hydration of slag. This is conditioned by the
hydraulic activity of slag which is influenced by its fineness, its
chemical composition and its glass contents and composition
[9,10].
El-Hadjar steel company, located in the East of Algeria, produces
a granulate slag, used in cement factories in substitution of clinker
with rates no greater than 20%. This low substitution rate is justi-
fied by the low reactivity of this Algerian slag. Previous studies
have shown a limited hydraulicity of this slag [11,12] even if the
chemical resistance of cement mixes containing this slag, espe-
cially for substitution rate exceeding 30%, was satisfactory. More-
over, it seems that, by increasing its fineness, it would be
possible to obtain good mechanical strengths. Nevertheless, only
few studies were conducted until now to point out its effect on
durability, particularly transport properties. These properties,

0950-0618/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2010.06.077
 A. Hadj-sadok et al. / Construction and Building Materials 25 (2011) 1018–1025
which mainly depend on the porosity and the microstructure, are
quite good indicators of the service life of concrete: all the mecha-
nisms of degradation are indeed depending on transportation pro-
cesses, either in gas or liquid phases [13].
This research project analyses the durability of cement mixes
where low reactive Algerian blast furnace slag is used.
2. Experimental program
2.1. Description of materials
Ordinary Portland cement (OPC) type CEM I 52.5N with a fineness of 4200 cm2/
g was used for mortar mixes. The ground granulated blast furnace slag (GGBFS)
used in this research project was produced in the iron steel company of El Hadjar
(Algeria). The GGBFS (Fig. 1) was ground in a laboratory mill to a Blaine fineness
of 4150 cm2/g. Its laser granulometry, with comparison to OPC, is presented on
Fig. 2. The chemical composition of cement and slag are given in Table 1. The slag
is characterized by low rate of alumina and magnesia, in comparison with an aver-
age slag chemical composition [14]. The hydraulic activity index [15], calculated on
the basis of the compressive strength, (Table 2) offers values from 0.54 at 7 days to
0.70 at 28 days: these values classify the ‘‘hydraulicity” of the slag as low to
moderate.
Standardized sand (EN 196) with maximum particle size of 2 mm has been used
for mortars mixes design.
2.2. Mix proportion, specimen preparation and curing
The specimens were a RILEM mortar (three parts of sand, one part of cement
and 0.5 part of water) in which cement has been partially substituted in mass with
0% (M0), 30% (M30) and 50% (M50) of slag, respectively.
All samples preparation was processed in a similar manner, according to Euro-
pean Standard EN 196-1 [16]. The mortars were cast into 4  4  16 cm moulds for
24 h and cured with plastic sheet. The hardened mortar was then demoulded and
kept at 20 ± 2 °C and under water until the age of testing.
2.3. Test procedure
2.3.1. Microstructure evaluation
Water porosity (Pw) was determined by water absorption tests on cylindrical
specimens (80 mm diameter and 40 mm length) cored from prismatic mortar slabs
(140  160  40 mm) after 28, 90 and 360 days curing, respectively. The specimen
is firstly dried up to 105 °C (less than 0.1% mass change in 24 h). After weighing,
specimens are immerged into water at 20 ± 2 °C until saturation [17].
In order to specify pore size distribution, the different mortars have been inves-
tigated using Mercury Intrusion Porosimetry (MIP). A Pascal 240 porosimeter was
used. The increasing step of pressure was fixed to 200 MPa, allowing investigation
of pore radius from 7.5 lm to 3.7 nm. The test was performed on approximately
2 cm3 mortar samples after drying at 50 °C and in presence of silica gel until con-
stant mass. The MIP test was carried out on 28, 90 and 360 days old samples,
respectively; for each mix, at least two samples were tested. The total porosity
PHG was measured as well.
Microstructure is observed using Scanning Electron Microscope (SEM). Samples
with a surface of 2 cm2 from each mix mortar were prepared and analyzed after
360 days curing.
Fig. 1. View at SEM of slag.
1019
OPC (CEM I- 52.5)
100
Cumulative passing (%)
Slag
80
60
40
20
0
0,01 0,1 1 10 100 1000 10000
Sieve diameter (µm)
Fig. 2. Laser granulometry of cement and slag.
2.3.2. Compressive strength
Three prismatic specimens (40  40  160 mm) have been moulded and cured
for compressive strength measurement. Compression strength is measured accord-
ing to EN 196-1 [16] after 2, 7, 28, 90 and 360 days, respectively.
2.3.3. Durability test
The durability performance was evaluated by means of two durability tests: (1)
water absorption by capillarity and (2) chloride diffusion.
For water capillary absorption test, cylindrical samples (80 mm diameter and
40 mm length) are dried in an oven at 45 ± 5 °C until constant weight (less than
0.1% mass change in 24 h with a minimum of 7 days). In order to ensure that water
flows in a single direction, specimens were sealed by epoxy resin on the side and
stored in water on their cross section with a constant water height of 2 mm. Water
absorption has been registered after 4, 9, 16, 36, 49, 64 min and 4, 8 and 24 h,
respectively. Sorptivity coefficient S is determined in accordance with EN 13057
[18].
Rates of diffusion of Cl ions into cement mortars are monitored using two
compartment diffusion cells (Fig. 3). Mortar blocks (8 mm thick) are sawed from
80 mm diameter specimens and stored in Ca(OH)2 saturated solution. Prior to the
test, each specimen is polished with 600-grade emery paper, rinsed with deionised
water and surface dried with a tissue before being fitted into the diffusion cell. After
fitting with epoxy resin and sealing with silicon paste, the cells are filled at one side
with a Ca(OH)2 solution and at the other side with 1 Mole NaCl in a saturated
Ca(OH)2 solution [19]. At periodic intervals, chloride concentration is determined
by potentiometric titration from a 10 ml sample of the solution in cell 2.
3. Results and discussion
3.1. Microstructure
3.1.1. Porosity
Total water porosity and MIP results are shown in Table 2 for
mortars with and without slag. The presence of slag seems to in-
crease the water porosity level ‘‘PW” of mortars at any age (28,
90 and 360 days). This effect is more pronounced for 50% slag mor-
tars. The porosity decreases with age for all mortars. However, the
decrease is more important between 90 and 360 days for slag mor-
tars (M30 and M50) compared to M0 mortar. Similar variations
were observed in the literature [4,20] and are mainly due to the
reduction of clinker content and to latent hydration of slag which
requires activation by Ca(OH)2 resulting from hydration of clinker.
The results of the MIP ‘‘PHG” show a similar variation versus
water porosity ‘‘PW”, even if lower values. This difference is related
to the difference of both methods of investigation: with Mercury
Intrusion Porosimetry, largest capillary pores, saturated at low
pressures, are not taken into account. On the other side, water
porosity considers the total open porosity of material.
3.1.2. Pore size distribution
Fig. 4 illustrates results of the pore size distribution for the
three types of mortars and its evolution between 28 and 360 days
wet curing. Medium radius pore (Rm) and threshold pore access ra-
dius (Rt) of mortars are presented in Table 3.
1020 A. Hadj-sadok et al. / Construction and Building Materials 25 (2011) 1018–1025

Table 1
Chemical composition of cement and El Hadjar GGBFS.

SiO2 CaO Al2O3 MgO Fe2O3 Free CaO Na2O K2O SO3 Loss of ignition Insoluble residue
OPC (CEM I 52.5) 18.40 61.30 5.60 0.90 3.80 0.20 0.42 0.78 3.30 2.20 0.50
GGBFS 41.20 42.84 9.19 2.12 3.44 – 0.10 0.70 0.15 0.20 –

Table 2
Compressive strength, water porosity PW and mercury porosity PHG of cement mortars with (M30 and M50) and without (M0) slag.

Compressive strength (MPa) Water porosity PW (% in volume) Mercury porosity PHG (% in volume)
Mortars Slag (%) 2 days 7 days 28 days 90 days 360 days 28 days 90 days 360 days 28 days 90 days 360 days
M0 0 24.3 35.3 58.9 68.6 76.0 16.47 15.99 15.72 11.20 10.90 10.40
M30 30 13.9 30.8 52.5 67.2 74.5 17.61 17.10 16.76 12.50 11.50 10.50
M50 50 6.6 19.0 41.0 55.7 67.5 18.98 18.31 17.90 14.15 13.45 11.60

A
4 4

Cell 1 Cell 2

Epoxy resin
A
1 2

1: saturated Ca(OH)2 solution + 1 M NaCl A-A

2: saturated Ca(OH)2 solution
3: specimen mortar
4: opening for collection of solution sample

Fig. 3. Experimental setup of diffusion cells.

At 28 days old, the pore distribution of slag mortars is coarser
than that of mortar without slag (M0). The threshold pore access
radius and the average radius of pores are higher in the mortars
M30 and M50 that in the M0 mortar. At 90 days old, the pore size
distribution of mortars with slag seems to evolve to smaller and fi-
ner pores until comparable porosity of M0 mortar: the threshold
pore access radius becomes indeed comparable for the three mor-
tars (approximately 36 nm), which is less important than at
28 days (38–40 nm), for all mortars studied. As for the average ra-
dius, it is less significant in the slag cement mortars.
In the long term, after 360 days, the pore structure of slag mor-
tars is clearly more refined than that of reference mortar (M0). This
effect is particularly visible for the mortar with 50% slag, where the
distribution of pores is mainly concentrated on radius values from
4 to 30 nm. This range is mainly related to the fine pore capillary
and the inter-crystallite porosity (between particles of gel hy-
drated) [5,21]. This observation is clearly in accordance with the
values of average pore radius, which is lower for M30 and M50
mortars compared to M0 mortar (Table 3).
However, it is interesting to remember that, even after
360 days, the mortars with slag (M30 and M50) have a volume of
pore higher than that of the M0 mortar. That could be due to the
higher porosity of the CSH resulting from the slag hydration [22].
Indeed, the increase of the pores volume for mortars with slag
(particularly M50) compared to M0 mortar (Fig. 4) is mainly in-
duced in the fraction of small radius pores (pore radius less than
35 nm and particularly between: 20 and 4 nm) this class of small
pores is usually attributed to the porosity of the CSH gel, which
is more present in the matrices containing slag cement [21,23].
3.1.3. SEM observation
Observation of M0 mortar sample in SEM reveals a structure of
hydrates (crystals and gel) combined to the ettringite needles
(Fig. 5a). In M50 mortar (Fig 5b), hydrates structure appears rela-
tively denser and ettringite needles are not detected: this could ex-
plain the porosity refinement observed in pore distribution of M50
mortar (MIP results) and thus reflects good hydration of slag after
360 days curing in wet conditions.
3.2. Compressive strength
The results of compressive strength development of mortars
containing GGBS as cement substitute material is given in Table
2. The addition of slag results in compressive strength decrease
for all ranges of substitution at 7 and 28 days, respectively. This
can be explained by the latent hydration of GGBFS and lower
hydraulic activity of slag used in this investigation. After 90 days
however, compressive strength of mortars with 30% slag was found
to be comparable to that of mortars without slag. But even after
360 days curing, strength reduction (in the order of 11%) is again
recorded for mortar containing 50% of slag; similar behaviour of
this Algerian slag was observed in previous researches
[11,12,24,25]. Fig. 6 plots the relationship between compressive
strength and total water porosity for all mixtures and all ages: an
 A. Hadj-sadok et al. / Construction and Building Materials 25 (2011) 1018–1025 1021

a 14 0,07

Cumulative pore volume (cm³/g)

M0 M30 M50

Incremental pore volume (%)

12 0,06
28 days
10 0,05

8 0,04

6 0,03

4 0,02

2 0,01

0 0,00
1 100 10000
Pores radius (nm)

b 14 0,07

Cumulative volume pore (cm³/g)

M0 M30 M50
Incremental pore volume (%)

12 0,06
90 days
10 0,05

8 0,04

6 0,03

4 0,02

2 0,01

0 0,00
1 10 100 1000 10000
Pores radius (nm)

c 14 0,07
M0 M30 M50

Cumulative pore volume (cm³/g)

Incremental pore volume (%)

12 0,06
360 days
10 0,05

8 0,04

6 0,03

4 0,02

2 0,01

0 0,00
1 10 100 1000 10000
Pores radius (nm)

Fig. 4. Distribution of capillary pore radius for different mortars at 28 day (a), 90 day (b) and 360 days (c).

Table 3
Threshold pore access radius (rt) and medium radius pore (rm) of cement mortar with 3.3. Durability tests
(M30 and M50) and without (M0) slag.

28 days 90 days 360 days 3.3.1. Capillary absorption

rt (nm) rm (nm) rt (nm) rm (nm) rt (nm) rm (nm)
Fig. 7 shows the capillary water absorption per unit area as a
function of square root of time for all mortar mixtures. At any
M0 38.0 35.0 36.0 32.0 31.5 30.5
M30 40.0 38.8 37.0 30.5 30.0 28.0
age (90 as 360 days), a low capillary absorption for the slag mortars
M50 40.0 35.8 36.0 27.5 28.0 25.3 compared to control mortar M0 is observed, particularly between 8
and 24 h absorption. However, the evolution of capillary absorp-
tion with age of maturation (between 90 and 360 days) remains
increase in compressive strength is associated with a decrease in low.
total porosity. A linear correlation (R = 0.80) is observed between Sorptivity coefficient S (Table 4) was calculated by the gradient
the two parameters. of the linear section (between 0 and 8 h) of the curves presented in
1022 A. Hadj-sadok et al. / Construction and Building Materials 25 (2011) 1018–1025

CEM I mortar 50% slag mortar

CEMI mortar 50% slag mortar

Fig. 5. SEM micrograph of (a) CEM I mortar and (b) 50% slag mortar after 360 days of wet curing.

Fig. 7. The presence of slag – particularly at 50% rates – induces
lower values of sorptivity coefficient: at 90 days old, in comparison
with M0 mortar, a decrease of S coefficient of 16% and 32% is re-
corded for M30 and M50 mortars, respectively. After 360 days,
the decrease reaches up to 35% for M50 mortar.
This improvement of the sorptivity can be explained once again
by the more refined pore structure of slag mortars. The relationship
between sorptivity coefficient and pores medium radius is plotted
in Fig. 8. A good correlation (R = 0.92) is proposed, showing a de-
crease of sorptivity proportional to the mean radius porosity. This
is in accordance with the Poiseuille law, as the capillary absorption
of a liquid is proportional to its radius [26].
The positive effect of slag on the sorptivity of mortars and con-
cretes has been mentioned in the literature not only for low slag
content [27] but especially for higher slag rates up to 50% [6,28].
Compressive strength (MPa)
The improvement is justified by the CSH gel, produced by the
hydration of slag [20] and modification of the distribution and
dimensions of capillary porosity.
3.3.2. Chloride diffusion
The evolution of chloride diffusion was measured during
270 days for mortars with and without slag and 90 days of wet cur-
ing. Fig. 9 shows the evolution of the concentration of chloride ions
in cell 2 versus time for the three mortars studied. The occurrence
time (breakthrough time) was calculated from the intercept of the
concentration versus time date (minimum rate of 30 mg/l of Cl
ions is reached in the cell 2). The results clearly show a higher dif-
fusion rate for M0 mortar in comparison with slag mortars: after
270 days, the concentration of chloride reaches about 1000 mg/l
for M0 compared to 370 mg/l and 40 mg/l for M30 and M50,
respectively (Fig. 9).
The effective diffusion coefficient (Table 5) is calculated when

75 the steady state is reached (Fig. 9), taking into consideration the
thickness and area of the mortar disk, the cell volume and the chlo-
65 ride concentrations. Chloride ions transfer through M0 specimens
is observed after only 45 days with an effective diffusion coeffi-
55 cient of 3.00  10 10 m2/s. For M30 mortar, the breakthrough time
is more important (73 days) with a coefficient 60% smaller than for
45
M0. Finally, for M50 mortar, steady state is not attained after
Experimental data 270 days diffusion, which does not allow quantifying diffusion
35
Resgression (R=0.80) coefficient. However, the breakthrough time of Cl ions is large
25
(between 210 and 240 days), which reflects a good resistance to
15 16 17 18 19 20 chloride penetration.
Total water porosity (%) Several researches [8,23,29] have also observed a decrease in
the chloride diffusion coefficient of mixtures containing slag ce-
Fig. 6. Correlation between compressive strength and water porosity. ment. The higher the rate of Ca(OH)2 in the hydrated matrix, the
 A. Hadj-sadok et al. / Construction and Building Materials 25 (2011) 1018–1025 1023

a 0.45
M0 M30 M50

Water absorption (g/cm²)

90 days

0.30

0.15

0.00
0 5 10 15 20 25 30 35 40
Time (min1/2)

b 0.45
Water absorption (g/cm²)

M0 M30 M50
360 days
0.3

0.15

0
0 5 10 15 20 25 30 35 40
Time (min1/2)

Fig. 7. Water absorbed per unit area of slag cement mortar at (a) 90 days and (b) 360 days.

Table 4
ture enhances the resistance to chloride penetration. The improve-
Sorptivity coefficient of mortars (kg/mm2 h1/2).
ment in diffusion is also attributed to the binding adsorption
Age M0 M30 M50 capacity of diffusing chlorides onto the hydrate slag walls (CSH
90 days 6.11 5.12 4.12 and CAH), which present a high specific surface [32,33], Dhir et
360 days 5.95 5.07 3.85 al. [34] and Luoa et al. [29] reported an increase in physical chlo-
rides binding when cement was partially substituted by slag.
In this research, it should be noted that mortars were cured in
35 wet conditions for 90 days before being placed in diffusion cells.
That means that the test period (9 months) favours continuous
Medium radius pore (nm)

hydration of slag allowing the densification of the matrix and pos-

sibly the chloride binding capacity of slag hydrates.
30

4. Conclusion

25 The objective of this research was to study the potential use of a

20
3 4
Sorptivity coefficient (kg/mm²/h1/2)
Fig. 8. Correlation between capillary coefficient and medium radius pore.
higher the volume of continuous pores [30]. Consequently, con-
sumption of Ca(OH)2 during the hydration of slag (pozzolanic reac-
tion) leads to a decrease in the volume of interconnected pores,
making the structure relatively more dense, tortuous and discon-
tinuous. On the other hand, the secondary CSH gel formed after slag
hydration, although less dense compared with primary CSH, is nev-
ertheless reported to be very effective in filling the larger capillary
pores in the matrix [31]. The resulting change in the microstruc-
Experimental data low reactive GGBFS from El Hadjar (Algeria) steel company. The
Regression (R=0.92) main conclusions to be drawn from a present experimental inves-
tigation are as following:
5 6 7
 Granulometry and specific surface of ground granulated blast
furnace slag is similar to Ordinary Portland Cement;
 the total porosity of slag mortars is higher than for OPC mortar;
 after 90 days, the pore size distribution of slag mortars is essen-
tially related to small pores (especially in the fraction of radius
less than 50 nm) leading to a densification of the microstruc-
ture. This effect is basically induced by a good hydration of slag
and is nevertheless more significant after a long period of wet
curing (360 days);
 compressive strength is lightly influenced by 30% substitution
of cement by GGBFS. However, 50% substitution led to a larger
reduction in strength, even after 90 days. This variation is
attributed to the moderate hydraulic activity of El Hadjar slag;
1024 A. Hadj-sadok et al. / Construction and Building Materials 25 (2011) 1018–1025

1,20

M0 M30 M50

Chloride concentration (g/l)

1,00

0,80

0,60

0,40

0,20

0,00
0 30 60 90 120 150 180 210 240 270 300
Time (days)

Fig. 9. Chloride diffusion rate for slag cement mortars.

Table 5
Chloride diffusion rate for mortars.
Mortars Breakthrough time (days) Effective diffusion coefficient (m2/s)
10
M0 45 3.00  10
10
M30 73 1.11  10
M50 210–240 – [12] Kenai S, Hadj Sadok A, Menadi B. Durability of concretes containing Algerian
 the sorptivity of mortar is improved by the presence of GGBFS,
especially for 50% substitution rate. However, a little evolution
of capillary absorption is observed between 90 and 360 days;
 chloride diffusion must be analyzed in terms of effective diffu-
sion coefficient but also of the breakthrough time. A substitu-
tion rate of 30% GGBFS induces an increase of the
breakthrough time, but with regards to OPC mortar, no major
differences are observed in diffusion coefficient. However, 50%
GGBFS led to better resistance of mortar to chloride penetration,
with a large breakthrough time and a low diffusing chloride
concentration.
These results are promising for the use of El Hadjar GGBFS in
concrete technology especially under aggressive environments
such as chloride attack and/or hot climate. The optimum substitu-
tion level seems to be around 30% to get good performance for both
mechanical and durability properties. Impact on rheology proper-
ties and mechanical performances of concrete should be now
investigated.
References
[1] Brand R, Pulles T, Van Gijlswijk R, Fribourg-Blanc B, Courbet C. European
pollutant emission register. Final report; 2004. <http://www.eper.cec.eu.int>.
[2] Regourd M. Cement Portland hydration – hydraulic concrete: knowledge and
practice. Paris: ENPC; 1995 [In French].
[3] Glasser FP. Properties of cement waste composite. Waste Manage
1996;6:159–68.
[4] Cheng A, Huang R, Jiann-Kuo W, Cheng-Hsin C. Influence of GGBS on durability
and corrosion behavior of reinforced concrete. Mater Chem Phys
2005;93(2):404–11.
[5] Ye G, van Breugel K. Simulation of connectivity of capillary porosity in
hardening cement-based systems made of blended materials. Concr Technol
2009;54(2/3):163–84.
[6] Guneyisi E, Gesoglu M. A study on durability properties of high-performance
concretes incorporating high replacement levels of slag. Mater Struct
2008;40(3):479–93.
[7] Shi HS, Xu BW, Zhou XC. Influence of mineral admixtures on compressive
strength, gas permeability and carbonation of high performance concrete.
Constr Build Mater 2009;23(5):1980–5.
[8] Yun Yeau K, Kyum Kim E. An experimental study on corrosion resistance of
concrete with ground granulate blast-furnace slag. Cem Concr Res
2005;35(7):1391–9.
[9] Neville AM. Properties of concrete. Paris: Eyrolles; 2000 [In French].
[10] Pal S, Mukherjee A, Pathak S. Investigation of hydraulic activity of ground
granulated blast furnace slag in concrete. Cem Concr Res 2003;33(9):1481–6.
[11] Kenai S, Amrane A. Mechanical properties and permeability of slag in Algerian
hot climate. In: Proceedings of the fourth international conference of concrete
technology for developing countries, Gazimagusa. North Cyprus; 1996.
pp.138–148.
slag; tenth international conference on non-conventional materials and
technologies, Cali, Colombia; 12–14 November, 2008.
[13] Baroghel-Bouny V. Design for specified life-time of concrete structures:
durability indicators. Paris: AFGC; 2004 [In French].
[14] Roper H. Composition, morphology, hydration properties and adhesive
properties of some granulated slags. In: Proceeding of 7th international
congress of cement chemistry. vol. 2. Paris; 1980. p. 13–17 [In French].
[15] NF P 18-506. Addition for hydraulic concrete, ground granulated blast furnace
slag. Paris: AFNOR; 1992 [In French].
[16] EN 196-1. European Committee for Standardisation. Methods of testing
cement. Part 1: determination of strength. Brussels; 1995.
[17] NBN B15-215. Water absorption of concretes; 1989.
[18] EN 13057. Products and systems for the protection and repair of concrete
structures – test methods – determination of resistance of capillary
absorption; 2002.
[19] Courard L, Darimont A, Schouterden M, Ferauche M, Willem X, Degeimbre R.
Durability of mortars modified with metakaolin. Cem Concrete Res
2003;33(9):1473–9.
[20] Jiang SP, Grandet J. Evolution comparée des porosités des mortiers de ciment
au laitier et des mortiers de ciment portland. Cem Concr Res 1989;19:487–95.
[21] Baroghel-bouny V. Cement pastes and concrete characteristics: methods,
analysis and interpretation. Paris: LCPC; 1994 [In French].
[22] Perlot C, Verdier V, Carcassès M. Influence of cement type on transport
properties and chemical degradation: Application to nuclear waste storage.
Mater Struct 2006;39(5):511–23.
[23] Han Young M, Hong Sam K, Doo Sun C. Relationship between average pore
diameter and chloride diffusivity in various concretes. Constr Build Mater
2006;20(9):725–32.
[24] Hadj Sadok A, Kenai S, Menadi B. Sulfate resistance of mortars with slag
cement. In: Proceedings of International symposium on characterisation and
modeling of materials and structures, paper GC 22, University of Tizi-ouzou,
Algeria; 2008 [In French].
[25] Behim M, Clastres P, Habita M, Carles-Gibergues A. Study of a new cement
with El Hadjar slag (Algeria). Annales du Bâtiment et des Travaux Publiques;
April 2004. p. 21–29 [In French].
[26] Courard L. Contribution to the analysis of parameters affecting the creation of
interface between concrete and repair systems. PhD thesis, faculty of applied
sciences, University of Liège, Belgium; 1999 [In French].
[27] Vejmelkova E, Pavlikova M, Keršner Z, Rovnanikova P, Ondracek M, Sedlmajer
M, et al. High performance concrete containing lower slag amount: a complex
view of mechanical and durability properties. Constr Build Mater
2009;23(6):2237–45.
[28] Alexander MG, Magee BJ. Durability performance of concrete containing
condensed silica fume. Cem Concr Res 1999;29(6):917–22.
[29] Luoa R, Caib Y, Wangb C, Huanga X. Study of chloride binding and diffusion in
GGBS concrete. Cem Concr Res 2003;33(1):1–7.
[30] Feldman RF. Significance of porosity measurements on blended cement
performance. In: Proceedings of 1st international conference on the use of
fly ash, silica fume, slag and other mineral products in concrete, vol. 1. ACI;
1983. p. 415–433.
[31] Wee TH, Suryavanshi AK, Tin SS. Evaluation of rapid chloride permeability test
(RCPT) result for concrete containing mineral admixture. ACI Mater J
2000;97(2):221–32.
 A. Hadj-sadok et al. / Construction and Building Materials 25 (2011) 1018–1025 1025

[32] Glasser FP, Marchand J, Samson E. Durability of concrete; degradation [34] Dhir RK, EL-Mohr MAK, Dyer TD. Chloride binding in GGBS concrete. Cem
phenomena involving detrimental chemical reactions. Cem Concr Res Concr Res 1996;26(12):1767–73.
2008;38(2):226–46.
[33] Arya C, Buenfeld NR, Newman JB. Factors influencing chloride binding in
concrete. Cem Concr Res 1990;20(2):291–300.