Construction and Building Materials 133 (2017) 468–481

Contents lists available at ScienceDirect

Construction and Building Materials

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

The use of the volcanic powders as supplementary cementitious

materials for environmental-friendly durable concrete
Y. Labbaci a, Y. Abdelaziz b,⇑, A. Mekkaoui c, A. Alouani c, B. Labbaci a
a
LMS Laboratory, Department of Civil Engineering, University of Bechar, 08000, Algeria
b
Department of Civil Engineering, University of Bechar, 08000, Algeria
c
FIMAS Laboratory, Department of Civil Engineering, University of Bechar, 08000, Algeria

h i g h l i g h t s

The powders of the volcanic rocks tested present a very high pozzolanic activity.

Relationships between SAI index and chemical components were established.

Energy consumption for the production of volcanic powder cements was calculated.

Volcanic powders can be used up to 15% as a partial substitute for Portland cement.

Mortars with the volcanic powder withstand much better chemical attacks.

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

Article history: This study is part of a sustainable development policy that is dictated by the growing needs of material
Received 17 June 2016 resources and the requirements of environmental protection. It addresses an investigation on the possi-
Received in revised form 22 October 2016 bility of using volcanic powders as supplementary cementitious materials for environmental-friendly
Accepted 20 December 2016
durable concrete. For this purpose, an experimental investigation was carried out to evaluate the
Available online 28 December 2016
mechanical resistances and the durability of mortars containing an amount of volcanic powders.
Firstly, the work attempts to characterize several volcanic rocks (basalt, olivine andesite, amphibole-
Keywords:
biotite andesite, amphibole andesite, hyodacite and scoria) from the mineral and chemical viewpoint
Volcanic powders
Cement
and evaluate their pozzolanic activity. Relationships between chemical components of volcanic rocks
Pozzolanic activity and strength activity index were established. Secondly, it considers the mechanical strengths of the
Compressive strengths Portland cement mortars containing a variable percentage of volcanic powders (10, 15, 20, 25, and
Durability 30%). Performance energy for the production of volcanic powder cements was evaluated. At the end,
the effects of aggressive chemical environments were investigated in terms of miscellaneous acidic
attacks (H2SO4, HCl, HNO3 and CH3COOH). Furthermore, a supplementary cementitious material, used
by many cement plants in Algeria, has been included in order to establish a comparative study.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction it possible substantially to decrease the gas emissions for purpose

of greenhouse (0.89t CO2/t clinker), minimize the harmful environ-
To lighten the economic and environmental impacts of cement mental impacts, and reduce the consumption of natural resources
industry, supplementary cementitious materials (SCMs), including and the energy. Due to the rush towards sustainability, nowadays,
limestone powder, fly ash, slag, silica fume, and natural pozzolans, many countries have adopted a sustainable development program
can be used as partial replacement of cement in concrete [1–4]. based on the use of the SCMs materials, namely natural pozzolanic
Apart from the immediate reduction consequence of the concrete materials. This subject has become of interest to the scientific com-
manufacturing costs, the partial replacement of cement will make munity; several researchers have focused their research efforts on
using natural pozzolan as an additive or substitute for cement in
concrete mixtures [5–8]. In fact, natural pozzolan would react with
⇑ Corresponding author. calcium hydroxide to form more cementitious calcium silicate
E-mail addresses: lyas6974@gmail.com (Y. Labbaci), abdelaziz970@yahoo.fr hydrate. In addition, mortars which contain pozzolanic materials
(Y. Abdelaziz), Dr.ySr2016@gmail.com (A. Mekkaoui), rdmrdm80@yahoo.fr
(A. Alouani), labonum@yahoo.fr (B. Labbaci).

http://dx.doi.org/10.1016/j.conbuildmat.2016.12.088
0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481 469

would exhibit considerable enhancement in durability properties The dimensions of this volcanic massif are approximately 12 km
[9–11]. long by 6 km wide (Fig. 1).
For this purpose, the aim of this investigation is to determine This massif displays a wide variety of magmatic products, such
the validity of volcanic powders used as supplementary cementi- as andesitic basalt flows, doleritic dykes and dacitic intrusions
tious materials (SCMs). This step will make it possible to develop (Fig. 2) [16–18]. The reserve of the volcanic rocks is a great poten-
materials available locally and will contribute to the social devel- tial for the cement industry; it is estimated at more than
opment of the area. It is important to note here that the results 150.106 m3 [19,20].
of this research are of special importance not only for studied The geological investigation in situ has directed us to choose
region but also for other regions with similar geology, such as: five representative samples (Fig. 3) showing separate textural char-
Ougarta, Yetti-Eglab and Edough (Algeria); Massif Central and acters (samples A, B, C, D and E). In addition to our investigation,
Alpes (France); Calabre (Italy); Sierra Nevada (Spain); Anti Atlas, we also established a comparative study by using supplementary
Ouarzazat, Ougnat and Siroua (Morocco); Harrat Al-Shaam (Syria, cementitious material (Sample F). This material is used as poz-
Jordan and Saudi Arabia); Djebel Dokhan (Egypt) etc. The choice zolans by local cement factories in the manufacture of composite
of this kind of materials has been recommended by their interest- Portland cement CPJ. A visual description of these tested samples
ing mechanical properties, abundance and economic exploitation is summarized in Table 1.
with lower environmental impact [12–14]. Indeed, from an energy
performance standpoint, the total energy required to produce 3. Materials and mix designs
cement was evaluated to be in the range 800–1200 kW h per ton
of cement, including around 50 kW h/t for the finish grinding of According to ASTM Standard C 311-98 [21], the volcanic rocks
the clinker. In the case of volcanic powder, the energy needed for were dried at a temperature of 105 °C to eliminate free water
the grinding of this material was evaluated at 75 kWh/t [15]. This and they were ground in a laboratory ball mill to a particle size
process energy was assumed to be zero kWh/t if the volcanic pow- of 100% less than 70 lm. The ordinary Portland cement type CEM
der is a by-product. I 42.5 obtained from cement factory of Ain Touta was used as
cementitious material. It had a Blaine specific area of 435 m2/kg
2. Location and geologic setting and a specific density of 3100 kg/m3. Standard sand of 2 mm
maximum particle size was employed for the formulation of
In this investigation, representative samples of volcanic rocks mortars. The distilled water was used in all mortar mixtures and
were shortlisted from natural deposits in the southwest of Algeria in the curing of all the tests specimens. The mix designs of mortars
(massif of Boukais). The massif of Boukaïs is located in the south- used for the experimentation are given in Table 2. After mixing, all
west of Algeria, specifically, at 50 km southwest of the Bechar city. mortars were cast into 4  4  16 cm molds and they were kept in
Its latitude and longitude are 31°900 –32°000 N and 2°400 –2°600 W. laboratory conditions for 24 h. Then, the specimens were

Fig. 1. Volcanic massif of Boukaïs. (A) Geographic location, (B) Panoramic illustration of the oasis of Boukaïs, (C) 3D illustration of the volcanic massif of Boukaïs.
470 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481

Fig. 2. Geological map of the volcanic massif [18].

demoulded and cured in lime-saturated water at 20 °C ± 2 until the Portland cement as follows: SAI = (rtm /rcm)  100, where rtm is
time of testing. the average compressive strength of test mixture specimen and
rcm is the average compressive strength of control mixture
specimen.
4. Experimentation In the third stage, the effect of using volcanic powder as partial
replacement of cement on compressive strength of mortars was
4.1. Methodology investigated. For this purpose, a party of ordinary Portland cement
was replaced by supplementary cementitious materials at 10%,
In the first stage, several volcanic rocks were characterized from 15%, 20%, 25% and 30%. Three samples 4  4  16 cm were tested
the mineral and chemical viewpoint. In order to make the petro- under third-point loading and six half-samples obtained were
graphic identification, a number of thin sections of volcanic rocks tested in uniaxial compression on a press controlled by force, in
were examined under a Nikon Polarized Light Microscope (Eclipse accordance with the EN 196-1 2005 standard [23]. A calculation
LV100Pol). The XRD powder pattern was recorded on a Siemens D- of the energy demand when volcanic powders are used as cement
5000 X-ray diffractometer, with Ka cobalt anticathode replacement was presented.
(k = 1.789 Å), at a current of 40 mA and voltage of 40 kV. The scans At the end of this investigation, the chemical durability of the
were performed in the 2h range from 0° to 70° with a scanning volcanic mixtures for aggressive chemicals was investigated in
speed of 2°/min. The chemical compositions of raw materials were terms of miscellaneous acidic attacks, namely: sulfuric acid (5%
determined by X-ray Fluorescence (XRF) using a Philips PW 1404 X H2SO4), hydrochloric acid (5% HCl), nitric acid (5% HNO3) and acetic
spectrophotometer. acid (5% CH3COOH). The samples 4  4  16 cm were made accord-
In the second stage, the evaluation of pozzolanic activity of var- ing to the standard ASTM C 267 2001 [24] and stored in water at
ious volcanic rocks by means of mechanical strength tests was 20 °C until the 28th day. Thereafter, the samples were weighed
established. The preparation of the mortars was carried out accord- to determine their masses before the chemical attack then they
ing to the norm NFP 15-403 [22]. A control mixture was produced were immersed in different aggressive mediums of inorganic and
with a constant binder/sand/water proportion of (1/3/0.5). In the organic acids. After each week the specimens were washed, dried
test mixtures, 20% of the mass of Portland cement used in the con- and weighed. The degree of chemical attack is measured by the fol-
trol mixture was substituted by the same mass of the test supple- lowing weight loss formula: Mass loss (%) = [(M1  M2)/M1)] 
mentary cementitious materials. The strength activity index with 100, were M1 is the mass of the specimen before immersion and
 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481 471

Sample A Sample B

Sample C Sample D

Sample E Sample F

Fig. 3. Tested samples.

Table 1
Visual description of tested samples. M2 is the mass of the specimen after immersion in chemical
solution.
Rock Color Special characteristics
A Dark –
B Reddish Small crystals of plagioclase 4.2. Results and discussion
Iron-rich matrix
C Gray Small crystals of plagioclase
D Gray – 4.2.1. Mineralogical and petrographical evaluations
E Greenish gray Large crystals of plagioclase The thin sections were analyzed to identify and quantify pri-
F Dark Striations due to flow structures mary minerals and alteration mineral assemblages, textures and
Very light weight due to numerous vesicles
structures (Fig. 4).

Table 2
Mortar proportions of mortars of the experimentation.

Portland cement [g] Volcanic powder [g] Graded standard sand [g] Water [ml]
CM 450 – 1350 225
VP10 405 45 1350 225
VP15 382.5 67.5 1350 225
VP20 360 90 1350 225
VP25 337.5 112.5 1350 225
VP30 315 135 1350 225

CM: Control mortar, VP: mortar using volcanic powder.

472 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481

A : Basalt B : Olivine andesite C : Amphibole- biotite andesite

D : Rhyodacite E : Amphibole andesite F : Scoria

Fig. 4. Microscopic observations under polarized light.LPA.25x. Pl: plagioclase, Ol: olivine, Cpx: clinopyroxene, Opx: orthopyroxene, Op: opaque Amp: amphibole, Bio: biotite,
Qz: quartz.

By means of the X-ray diffraction patterns, the constituents of reddish-brown color. In thin section, the sample shows a micro-
raw materials were identified (Figs. 5–10). The mineralogical iden- litic porphyritic texture. It contains mainly phenocrysts of fresh
tification by X-ray diffraction is summarized in Table 3. augite, sericitized andesine and minor amounts of altered
The mineralogical and petrographic evaluation of samples olivine and orthopyroxene.
shows a wide range of igneous rock compositions. It can be noted
The sample C corresponds to an amphibole-biotite andesite
that the tested volcanic rocks could be generally described as small crystals of feldspars. Under the microscope, it shows a
material with porphyritic microstructure texture, except the sam- microlitic porphyritic texture porphyritic microstructure in
ple E which has a hyalo-microlitic texture. The main mineral con- which sericitized andesine phenocrysts are zoned, together
stituents of these rocks are plagioclase and alkali-feldspar, with orthoclase, and including biotite, amphibole and apatite.
clinopyroxene and oxide minerals (magnetite and ilmenite).
The sample D corresponds to a rhyodacite. It is light gray in
According their microstructure and modal composition, the tested color. In thin section, the sample shows a hyalo-microlitic tex-
volcanic rocks can be subdivided into five kinds: ture. It contains some porphyritic feldspar, quartz and mafic.
It is composed of calcium feldspar porphyritic compounds, fer-

The sample A corresponds to a basalt. It is black to gray in color. romagnesian pyroxene, amphibole and biotite. Quartz is auto-
It has a microlitic porphyritic texture. Plagioclase is abundant in morphe embedded in a matrix hyalo-microlitic. Feldspars are
microphenocrysts and microlites. The clinopyroxene occur as partially or completely replaced by secondary minerals group
large twinned crystals. Plagioclase occurs in smaller crystals. phyllosilicates (sericite and chlorite) and calcite. Ferromagne-
Opaque are ferrotitanium oxides. sian are partially replaced by oxides and chlorite.

The sample B is an olivine andesite. It is reddish in color.
The sample E represents an amphibole andesite with large crys-
It has small crystals of plagioclase and an iron-rich matrix. tals of plagioclase. It has a greenish gray color. Under the micro-
Much disseminated, dustlike red iron oxide gives the rock a scope this andesite shows microlitic porphyritic texture. It
 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481 473

Fig. 5. X-ray diffraction data of basalt (Sample A).

Fig. 6. X-ray diffraction data of olivine andesite (Sample B).

consists essentially of plagioclase in phyric and microliters and
The sample F represents a scoria. It has a black color. It consists
rare amphibole crystals. Cavities filled with secondary quartz of amorphous glassy ground mass, vesicles, plagioclase and oli-
are also observed. vine [11].
474 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481

Fig. 7. X-ray diffraction data of amphibole- biotite andesite (Sample C).

Fig. 8. X-ray diffraction data of rhyodacite (Sample D).

4.2.2. Chemical composition the requirements of stipulated by the ASTM C 618 standard. The
The Table 4 gives the results of chemical composition of cement, tested volcanic powders (basalt, olivine andesite, amphibole-
volcanic powders [18]. They were compared with the ASTM C618- biotite andesite, rhyodacite, amphibole andesite and scoria) have
00 requirements [25] in Table 5. As seen from Tables 4 and 5, the a sum of SiO2, Al2O3, Fe2O3 equal to 76.67, 81.68, 77, 82.13, 81.15
chemical composition of all volcanic powders conforms well to and 76.05%, respectively. These values are greater than chemical
 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481 475

Fig. 9. X-ray diffraction data of amphibole andesite (Sample E).

Fig. 10. X-ray diffraction data of scoria (Sample F).

requirements according to ASTM C618 [25] (summary of SiO2 and 36.37% which are beyond chemical requirements (34%). All tested
Al2O3 and Fe2O3 higher than 70%). MgO, SO3 and loss on ignition volcanic powders show a strong acidic character, having a (SiO2 +
content are lower than the requirement percentage (5%, 4% and Al2O3) content ranging around 66.06–80.74% of the total. The
10% respectively). It can also be seen that the difference between importance of this content is emphasized by the fact that the active
SiO2 and CaO is equal to 42.49, 54.34, 45.76, 61.49, 51.25 and vitreous phases generally are richer in silica and alumina content.
476 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481

Table 3 Strength acvity index [%]

Mineralogical identification by X-ray diffraction of raw materials. 100
Sample Rock name Minerals 90
code 80
70
Reference
A Basalt Chlorite, calcite, labradroite, augite,
hematite, quartz 60
B Olivine andesite Chlorite, calcite, labradroite, augite, hematite 50
C Amphibole-biotite Quartz, hematite, chlorite, kaolinite, illite, 28 days
40
andesite labradroite, augite
D Rhyodacite Calcite, augite, chlorite, quartz, illite, 30
hematite, kaolinite, labradroite 20
E Amphibole Chlorite, hematite, quartz, labradroite, 10
andesite kaolinite, augite, illite, calcite 0
F Scoria Labradorite, augite, montmorillonite, A B C D E F
kaolinite, hematite, analcime
Fig. 11. Pozzolanic reactivity of the additions at 28 days.

From the chemical point of view, all the tested volcanic powders 91.90% and 84.15%, respectively). Among these samples, rhyo-
(basalt, olivine andesite, amphibole-biotite andesite, amphibole dacite powder gives the most reactive pozzolan (91.90%).
andesite, hyodacite and scoria) are suitable for use as supplemen- In order to make the test results interpretation easier, relation-
tary cementitious materials. ships between strength activity index and chemical components
were evaluated. From the correlation graph it is clear that there
is a significant correlation between the chemical components and
4.2.3. Evaluation of strength activity index the strength activity index (Fig. 12). Test results revealed that the
According to ASTM C595-08 [26], a pozzolan is defined as ‘‘a pozzolanic activities of volcanic rocks are directly proportional
siliceous or siliceous and aluminous materials which in itself pos- with the increase of the sum of (SiO2, Al2O3 and Fe2O3). It also
sesses little or no cementitious value but in the presence of mois- appears from the results that the pozzolanic activity increases with
ture, chemically react with calcium hydroxide at ordinary the increase in silicon dioxide rate (SiO2). It can be noted that
temperature to form compounds possessing cementious proper- Aluminium oxide (Al2O3), Iron (III) oxide (Fe2O3), LOI, and Magne-
ties”. It is known that its pozzolanic activity depends on a number sium oxide (MgO) decrease the pozzolanic activity.
of factors, the most significant of which seem to be the mineralog-
ical and chemical composition of the additive, extent of the amor- 4.2.4. Influence of replacement level
phous phase, specific surface area, content of Ca(OH)2 in the At this stage, an experimental investigation was carried out to
cement paste, mixing conditions and water/cement [27–29]. evaluate the mechanical resistances of mortars containing a vari-
The test results of the strength activity index at 28 days are able percentage of volcanic powders (10, 15, 20, 25 and 30%).
given in Fig. 11. According to ASTM C618-00 [25], for natural poz- Two kinds of volcanic powders were studied: amphibole andesite
zolan at a cement replacement of 20%, it requires a strength activ- and scoria. The choice of this first type of volcanic powders has
ity index greater than 75% at 28 days. The study of pozzolanic been recommended by its abundance and high pozzolanic activity.
activity provides that tested specimens (basalt, olivine andesite, Furthermore, a supplementary cementitious material ‘‘scoria”,
amphibole-biotite andesite, amphibole andesite and rhyodacite) used as pozzolans by local cement factories, has been included in
were all highly pozzolanic at 28 days (79.17%, 89.32%, 89.79%, order to establish a comparative study.

Table 4
Chemical composition of raw materials.

Cement A B C D E F
SiO2 (%) 21.04 49.97 56.26 51.88 67.21 57.15 47.21
Al2O3 (%) 5.34 16.64 14.15 16.31 13.53 16.28 18.85
Fe2O3 (%) 4.07 10.06 11.27 8.81 5.96 7.72 9.99
MnO (%) – 0.12 0.12 0.14 0.05 0.10 –
MgO (%) 1.14 4.14 3.26 3.98 1.05 3.75 4.38
CaO (%) 64.94 7.48 1.92 6.12 0.57 5.9 10.84
Na2O (%) 0.17 3.52 4.87 3.93 3.48 3.81 0.81
K2O (%) 0.88 2.37 2.71 3.09 5.44 0.99 0.20
SO3 (%) 0.75 – – – – – 0.50
TiO2 (%) – 1.81 2.23 1.62 0.42 1.08 –
P2O5 (%) – 0.36 0.37 0.34 0.08 0.23 –
LOI (%) 1.67 2.54 1.97 2.38 1.4 2.89 3.91

Table 5
Comparison between supplementary cementitious materials properties and ASTM C618-00 requirements.

Chemical composition Requirement Common NP (%) A B C D E F

SiO2 + Al2O3 + Fe2O3 Minimum 70 76.67 81.68 77 82.13 81.15 76.05
MgO Maximum 5 4.14 3.26 3.98 2.06 3.75 4.38
SO3 Maximum 3 – – – – – 0.50
Loss on ignition Maximum 10 2.54 1.97 2.38 1.4 2.89 3.91
SiO2-CaO Minimum 34 42.49 54.34 45.76 61.49 51.25 36.37
 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481 477

Fig. 12. Relationships between chemical components and strength activity index.
Relave compressive strengths (%)

120
Relave compressive strengths (%)

100
90
100
80
80 70
10% 10%
60
60 15% 50 15%
20% 40 20%
40 30
25% 25%
20
20 30% 30%
10
0 0
2 days 7 days 28 days 56 days 90 days 2 days 7 days 28 days 56 days 90 days
Age (days) Age (days)

Fig. 13. Compressive strength development of different mortars using amphibole Fig. 14. Compressive strength development of different mortars using scoria as
andesite as supplementary cementitious materials. supplementary cementitious materials.
478 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481

The relative compressive strengths development of different 50

batches at a different age is graphically illustrated in Figs. 13 and
14. The results reported are the average of six compression tests 40
at 2, 7, 28, 56 and 90 days.
According the results given by Figs. 13 and 14, as expected, all 30

Mass loss (%)

mortars, with and without volcanic powder show an increase in
strength with curing time. This ascertainment may be attributed CM
20
to hydration of cement which generates changes in the compact- Andesite
ness. Also, it is seen that as the volcanic powder replacement level
10 Scoria
increases as the compressive strength of volcanic powder-based
cement mortars decreases for all curing times. Specially, we noted
that, at an earlier age, the compressive strength decreases signifi- 0
cantly with increasing percentage of volcanic powder. This reduc- 1 day 7 days 14 days 21 days 28 days
tion in the compressive strength is attributed mainly to slower -10
pozzolanic activity at room temperature of the supplementary Immersion time (days)
cementitious materials [30,31]. This ascertainment is explained Fig. 17. Loss of mass of mortars at different ages of immersion in sulfuric acid (5%
by the interaction between the reactive silica which is in the glassy H2SO4).
portion of the addition and the Ca (OH)2 released by the hydration
of the cement. However after a few weeks, the compressive
strengths were increased. Indeed, at 28 days, a compressive 50
strength comparable to that of the control mortar was observed 45
for mortars containing 10% of andesite-based cement. At 56 days,
40
an increase in compressive strengths was observed for mortars
containing 15% and 20% andesite-based cement. The volcanic pow- 35

Mass loss (%)

der improves compressive strength so long as it gives rise to a sec- 30
ond C–S–H. During the hydration of cement, the reaction between 25 CM
Ca (OH)2 liberated and volcanic powder led to a refinement of the 20 Andesite
pore structure resulting in a highly impermeable matrix. Conse- 15
quently, the mortars containing volcanic powders are of compres- Scoria
10
sive strengths comparable with those of the control mortar starting
from 56 days on. 5
In the case of mortars containing scoria-based cement, it is pos- 0
1 days 7 days 14 days 21 days 28 days
sible to note that, at all levels of replacement at all ages, scoria
Immersion time (days)

Fig. 18. Loss of mass of mortars at different ages of immersion in hydrochloric acid
(5% HCl).
1000
Energy consumption (kWh/t)

950

900 50
45
850
40
800 Volcanic Powder
35
Mass loss (%)

750 Cement 30
700 25 CM
20 Andesite
650
0 10 20 30 40 15 Scoria
Volcanic powder content (%) 10

Fig. 15. Energy consumption for the production of one ton of mortar.
5
0
1 days 7 days 14 days 21 days 28 days
24 Immersion time (days)
Energy consumpon (kWh/MPa)

23 Fig. 19. Loss of mass of mortars at different ages of immersion in nitric acid (5%
22 HNO3).

21
20
19 Volcanic Powder induced compressive strength reductions in mortars. These
obtained results are in perfect agreement with those already found
18 Cement
by Senhadji [32]. The authors noted that the mortars containing
17
the scoria exhibit compressive strengths comparable to those of
16 the control mortar starting from the period of 90 days.
15
0 10 20 30 40
Volcanic powder content (%) 4.2.5. Calculation of the energy consumption
On the basis of data reported in the literature [33–35], a simple
Fig. 16. Performance energy for the production of one ton of mortar. calculation of the energy demand when volcanic rocks are used as
 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481 479

50 The Figs. 15 and 16 give the calculation results of the energy

45 consumption and the performance energy, respectively. It can be
seen that the energy consumption for the production of one ton
40
of volcanic powder cements was smaller than to that of cement
35
Mass loss (%)

alone for miscellaneous volcanic powder contents (10, 15, 20, 25

30 and 30%). From an energy performance viewpoint, volcanic powder
25 CM can be qualified as environmentally acceptable in the case of 15% of
20 Andesite volcanic powder. This content led to a better result than the refer-
15 ence in terms of energy consumption relative to the performance
Scoria
of the mortars.
10
5 4.2.6. Durability of mortars against acidic attacks
0 The test results of mass loss versus time for mortar samples
1 days 7 days 14 days 21 days 28 days
exposed to miscellaneous chemical attacks (5% H2SO4, 5% HCl, 5%
Immersion time (days) HNO3 and 5% CH3COOH) are shown in Figs. 17–20.
Fig. 20. Loss of mass of mortars at different ages of immersion in acetic acid (5%
Compared to sample kept in water, the state of samples after
CH3COOH). 28 days of immersion in miscellaneous chemical solutions (5%
H2SO4, 5% HCl, 5% HNO3 and 5% CH3COOH) is given in Fig. 21.
The Figs. 22 and 23 give the gain in mass of mortars with addi-
cement replacement can be evaluated. The energy consumption for tions (andesite and scoria) compared to control mortar after
the production of one ton of mortar is evaluated by using the fol- 28 days of immersion in acidic solutions (sulfuric acid, hydrochlo-
lowing relation [15]: ric acid, nitric acid and acetic acid).
Through these tests of chemical attacks, it is possible to high-
E ¼ Cc Ec þ Cvp Evr
light the following key findings:
where: E is the energy consumption for the production of one ton of
mortar, Cc and Cvp are the proportions of cement and volcanic pow-
All the mortars immersed in the aggressive solutions present a
der in the mortar, Ec is the energy consumption of the cement man- permanent loss of the mass as a function of the immersion per-
ufacturing process (950 kWh/t) and the finish grinding of the iod. However, the speed of the mass losses differs from one case
cement (50 kWh/t), Evr is the energy consumption for the grinding to another following the mortar composition, the chemical solu-
of the volcanic rocks (75 kWh/t). tion nature and the immersion time. All these attacks on mor-
The performance energy can be calculated by dividing the tars are associated with erosion and softening due to leaching
required energy by the compressive strength of the mortar. of Ca and decalcification of C–S–H [36].

Fig. 21. State of the samples after 28 days of immersion in acidic solutions compared to sample kept in water.
480 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481

100 5. Conclusion
90
80 On the basis of this experimental investigation, it is possible to
70 64.58 put forward the following principal conclusions:
60 54.38
50
The mineralogical and petrographic evaluation of samples
40 shows a wide range of igneous rock compositions. According
30 their microstructure and modal composition, the tested vol-
19.68 19.9
20 canic rocks can be subdivided into five kinds: basalt, olivine
10 andesite, amphibole-biotite andesite, amphibole andesite and
0 rhyodacite.
0
Control mortar sulfuric acid hydrochloric nitric acid acec acid
The conformity of the chemical composition of the volcanic
acid powders tested with the regulations required by standard ASTM
C618-00. It was found that the total silicon, aluminium and iron
Fig. 22. Gain in mass of mortars containing andesite-based cement compared to
control mortar after 28 days of immersion in acidic solutions.
oxide content is higher than the minimum requirement. Sulfur
trioxide, alkalis and loss on ignition are much lower than the
upper limit of the ASTM C618-00.

The powders tested present a very high pozzolanic activity in
perfect compliance with standard ASTM C618-98. The strength
100
activity index with portland cement ranging from 82.11% to
90
91.91% at 28 days old, which is over 75% as specified by ASTM
80
standard C 618-98.
70 65.96

Among the five analyzed volcanic powders, rhyo-dacite rock
60 54.15 presents the highest pozzolanic activity. It have been noted also
50 that the pozzolanic activity of the volcanic powders (olivine
40 andesite, amphibole-biotite andesite, amphibole andesite and
30 24.29 rhyodacite) is higher than that of BS’s pozzolan (scoria) used
20 14.5 by many local cement factories.
10
Relationships between strength activity index and chemical
0 components of volcanic rocks were established. Correlation
0
Control mortar sulfuric acid hydrochloric nitric acid acec acid results revealed that the pozzolanic activities of volcanic rocks
acid are directly proportional with the increase in silicon dioxide
rate (SiO2). Conversely, an increase in Al2O3, Fe2O3 and MgO
Fig. 23. Gain in mass of mortars containing scoria-based cement compared to
control mortar after 28 days of immersion in acidic solutions.
content of volcanic rocks adversely affected the pozzolanic
activity.

Mortars containing andesite-based cement withstand much
better acidic attacks compared to the control mortars. This

There is a positive effect when inserting the mineral additions resistance is considered to be comparable to the results of mor-
on the improvement of the chemical resistance of mortars tars containing scoria-based cement.
andesite and scoria with respect to the attacks of acids. This
Based on the compressive mechanical and chemical results
positive effect can be attributed to the creation calcium alumi- obtained, it is suggested that volcanic powders can be used up
nate hydrates; which are chemically more stable than the cal- to 15% as a partial substitute for Portland cement. This content
cium silicate hydrates in very acidic environments [37]. led to a better result than the reference in terms of energy con-

In the case of a chemical attack by sulfuric acid, a weight gain sumption relative to the mechanical performance and long-
was observed in the first days. This gain is due to the formation term durability of the mortars.
of calcium sulfate following the reaction of the calcium hydrox-
ide with the sulfuric acid Ca(OH)2 + H2SO4 ? CaSO4 + 2H2O. According to the results of this research, it can be concluded
Deposited gypsum forms a thick white layer on the surface of that volcanic powders can be used as effective pozzolanic materials
the mortars. While the chemical attack continues, the gypsum to replace portion of Portland cement. They allow a high compres-
layer (calcium sulfate) precipitated is easily leached, resulting sive strength and good resistance to various chemical attacks.
in a considerable mass loss. In addition, the calcium sulfate Moreover, the use of this material in its natural state, without cal-
formed in the initial operation reacts with the calcium cination, will allow an economic exploitation with lower environ-
aluminate phase in cement to form the hydrated sulfoalu- mental impact. At the end, this study can be regarded as a first step
minate calcium (Ca6 Al2(SO4)3(OH)1226H2O) ‘‘ettringite”. The in the valuation of the volcanic powders. However, much effort
hydrated sulfoaluminate calcium is an expansive element must be made to achieve this goal.
which creates cracks, loss of cohesion and strength and changes
in porosity.
References

According to above results, it is found that the mortars contain-
ing andesite-based cement exhibit better resistivity relative to [1] K. Yang, Y. Jung, M. Cho, S. Tae, Effect of supplementary cementitious materials
the reference mortar. It is noted, mainly, that after 28 days of on reduction of CO2 emissions from concrete, J. Clean. Prod. 103 (15) (2015)
774–783.
immersion, that the addition of volcanic powders reduces the
[2] B. Lothenbach, K. Scrivener, D. Hooton, Supplementary cementitious materials,
mass loss, compared to the reference mortar, 19.68% in sulfuric Cem. Concr. Res. 41 (12) (2011) 1244–1256.
acid, 64.58% in hydrochloric acid, 54.38% in nitric acid, and [3] H. Owaid, B. Roszilah, M. Taha, A review of sustainable supplementary
19.9% in acetic acid. These results are considered to be compa- cementitious materials as an alternative to all-portland cement mortar and
concrete, Aust. J. Basic Appl. Sci. 6 (9) (2012). 2887–303.
rable to the results of the mortars containing scoria-based [4] G.K. Al-Chaar, M. Al-Kadi, P.G. Asteris, Natural pozzolan as a partial substitute
cement. for cement in concrete, Open Constr. Technol. J. 7 (2013) 33–42.
 Y. Labbaci et al. / Construction and Building Materials 133 (2017) 468–481 481

[5] N. Kaid, M. Cyr, S. Julien, H. Khelafi, Durability of concrete containing a natural [21] ASTM Standard C 311-98, Standard Test Methods for Sampling and Testing Fly
pozzolan as defined by a performance-based approach, Constr. Build. Mater. 23 Ash or Natural Pozzolans for Use in Portland-Cement Concrete, ASTM
(12) (2009) 3457–3467. International, West Conshohocken, 2007.
[6] Y. Senhadji, G. Escadeillas, M. Mouli, H. Khelafi, A. Benosman, Influence of [22] NFP 15-403, Normal and Normal Mortar Sand, AFNOR, 2006.
natural pozzolan, silica fume and limestone fine on strength, acid resistance [23] NF EN 196-1, Methods of Testing Cement. Determination of Strength, AFNOR,
and microstructure of mortar, Powder Technol. 254 (2014) 314–323. 2005.
[7] N. Kaid, M. Cyr, H. Khelafi, Characterization of an Algerian natural pozzolan for [24] ASTM C 267-96, Standard Test Methods for Chemical Resistance of Mortars,
its use in eco-efficient cement, Int. J. Civ. Eng. 13 (4) (2015) 444–454. Grouts, and Monolithic Surfacings and Polymer Concretes, ASTM International,
[8] Z. Makhloufi, S. Aggoun, B. Benabed, E. Kadri, M. Bederina, Effect of magnesium West Conshohocken, 2001.
sulfate on the durability of limestone mortars based on quaternary blended [25] ASTM Standard C618-00, Specification for Coal Fly Ash and Raw or Calcined
cements, Cem. Concr. Compos. 65 (2016) 186–199. Natural Pozzolan for Use as a Mineral Admixture in Portland Cement Concrete,
[9] M. Ghrici, S. Kenai, M. Said-Mansour, Mechanical properties and durability of ASTM International, West Conshohocken, 2000.
mortar and concrete containing natural pozzolana and limestone blended [26] ASTM C595-08, Standard Specification for Blended Hydraulic Cements, ASTM
cements, Cem. Concr. Compos. 29 (7) (2007) 542–549. International, West Conshohocken, 1998.
[10] R.E. Rodriguez-Camacho, R. Uribe-Afif, Importance of using natural pozzolans [27] A. Shvarzman, K. Kovler, I. Schamban, Shter.G. Grader, Influence of chemical
on concrete durability, Cem. Concr. Res. 32 (2002) 1851–1858. and phase composition of mineral admixtures on their pozzolanic activity,
[11] A. Al-Swaidani, S. Aliyan, Effect of adding scoria as cement replacement on Adv. Cem. Res. 14 (2002) 35–41.
durability-related properties, Int. J. Concr. Struct. Mater. 9 (2) (2015) 241–254. [28] V. Papadakis, S. Tsimas, Supplementary materials in concrete. Part I: efficiency
[12] L. Yu, S. Zhou, W. Deng, Pozzolanic activity of volcanic rocks from Southern, and design, Cem. Concr. Res. 32 (2002) 1035–1041.
Jiangxi Province, China, J. Sustain. Cem. Based Mater. 5 (1) (2015) 176–198. [29] A. Çavdar, S ß . Yetgin, Availability of tuffs from northeast of Turkey as natural
[13] M. Çullu, H. Bolat, A. Vural, E. Tuncer, Investigation of pozzolanic activity of pozzolan on cement, some chemical and mechanical relationships, Constr.
volcanic rocks from the northeast of the Black Sea, Sci. Eng. Compos. Mater. 23 Build. Mater. 21 (2007) 2066–2071.
(3) (2016) 315–323. [30] M. Shannag, A. Yeginobali, Properties of pastes, mortars and concretes
[14] I. Uzun, S. Terzi, Evaluation of andesite waste as mineral filler in asphaltic containing natural pozzolan, Cem. Concr. Res. 25 (1995) 647–657.
concrete mixture, Constr. Build. Mater. 31 (2012) 284–288. [31] S. Mindess, J.F. Young, D. Darwin, Concrete, 2nd ed., Prentice Hall, Upper
[15] M. Hamidi, L. Kacimi, M. Cyr, P. Clastres, Evaluation and improvement of Saddle River, NJ, 2003.
pozzolanic activity of andesite for its use in eco-efficient cement, Constr. Build. [32] Y. Senhadji, G. Escadeillas, H. Khelafi, M. Mouli, A. Benosman, Evaluation of
Mater. 47 (2013) 1268–1277. natural pozzolan for use as supplementary cementitious material, Eur. J.
[16] M. Meindre, Geology of massive Boukaïs, Sci. Econ. Bull. 2 (1952) 8–45. Mining Environ. Civ. Eng. 16 (1) (2012) 77–96.
Research Office of Algeria. [33] N. Madlool, R. Saidur, M. Hossain, N. Rahim, A critical review on energy use
[17] A. Seddiki, Petrological and Geochemical Study of Magmatic Rocks of Boukaïs and savings in the cement industries, Renew. Sustain. Energy Rev. 15 (4)
(Northwestern Bechar, Algeria sudoccidentale). Contribution to a Geodynamic (2011) 2042–2060.
Study of the Region to the Infra-Cambrian (PhD thesis), University of Science [34] I. Bribián, A. Capilla, A. Usón, Life cycle assessment of building materials:
and Technology, 1997. comparative analysis of energy and environmental impacts and evaluation of
[18] A. Seddiki, N. Remaci-Benaouda, J. Cottin, B. Moine, R. Ménot, C. Perrache, The the eco-efficiency improvement potential, Build. Environ. 46 (5) (2011) 1133–
volcano-sedimentary Boukais inlier (south-western Algeria): evidence for 1140.
lithospheric thinning during the Late Neoproterozoic, J. Afr. Earth Sci. 39 [35] C. Hosten, B. Fidan, An industrial comparative study of cement clinker grinding
(2004) 257–266. systems regarding the specific energy consumption and cement properties,
[19] A. Zerrouki, Lithostratigraphy of the Proterozoic Boukaïs region and study of Powder Technol. 221 (2012) 183–188.
associated mineralization (Bechar, south-western Algeria), Geol. Surv. Algeria [36] H.T. Cao, I. Bucea, A. Ray, S. Yozghatlian, The effect of cement composition and
11 (2) (2000) 163–183. pH of environment on sulfate resistance of Portland cements and blended
[20] K. Tazghout-Graine, Geodynamics and Metallogenic Framework of Massif of cements, Cement Concr. Compos. 19 (2) (1997) 161–171.
Boukaïs and Mountains Ougarta (Bechar/Southwest Algeria) (PhD thesis), [37] E. Hewayde, M. Nehdi, E. Allouche, G. Nakhla, Using concrete admixture for
University of Science and Technology, 2010. sulfuric acid resistance, ICE Proc. Constr. Mater. 160 (1) (2007) 25–35.