Construction and Building Materials: Dimitra Zaharaki, Michalis Galetakis, Kostas Komnitsas

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Construction and Building Materials 121 (2016) 686–693

Contents lists available at ScienceDirect

Construction and Building Materials


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

Valorization of construction and demolition (C&D) and industrial wastes


through alkali activation
Dimitra Zaharaki, Michalis Galetakis, Kostas Komnitsas ⇑
School of Mineral Resources Engineering, Technical University of Crete, University Campus, 73100 Chania, Greece

h i g h l i g h t s

 Waste co-valorization is in line with zero-waste and circular economy principles.


 Molar ratios in raw materials, such as SiO2/Al2O3 and SiO2/CaO affect process.
 Final products exhibit good strength and thermal properties.

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

Article history: In the present experimental study, the valorization of construction and demolition (C&D) and industrial
Received 3 March 2016 wastes, namely ferronickel slag and red mud, through alkali activation is investigated. Specimens were
Received in revised form 13 June 2016 produced by mixing various proportions of the raw materials with NaOH and sodium silicate solutions.
Accepted 14 June 2016
The paste was cast in cubic metal moulds with an edge of 5 cm, cured at 80 °C for 24 h and then aged for
Available online 20 June 2016
7 days at room temperature. The produced specimens were subjected to compressive strength testing.
The effect of the molarity of the activating solution, the mineralogy of the raw materials and the ratios
Keywords:
of SiO2/Al2O3 and SiO2/(Al2O3 + CaO) on the compressive strength of the final products was investigated.
Alkali activation
Geopolymer
Also, the effect of high temperature heating (400–800 °C) on the structural integrity of the produced
Slag specimens was assessed. The use of analytical techniques, namely X-ray diffraction,
Construction and demolition wastes Thermogravimetric Analysis, Fourier Transform Infrared Spectroscopy and Scanning Electron
Red mud Microscopy provided insights on the morphology, structure and thermal resistance of the produced
specimens.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction of hazardous elements. Recent research efforts have mainly


focused on the valorization of a wide variety of wastes and the pro-
Alkali activation of aluminosilicates at relatively low tempera- duction of materials with appropriate physico-chemical, mechani-
ture results in the production of cementitious materials, usually cal and thermal properties [5–10].
called geopolymers or inorganic polymers, which are characterized The reactivity of raw materials during alkali activation is mainly
by a partially or fully amorphous polymeric structure consisting of related to the percentage of their amorphous content [11]. The
Si–O–Al bonds. Their properties such as high early strength, chem- mechanisms involved in this process have been elucidated to a cer-
ical and high temperature resistance, depend mainly on the miner- tain extent by the use of several analytical techniques including
alogy of the raw materials as well as on the strength of the alkaline Scanning Electron Microscopy (SEM), Fourier Transform Infrared
activator used [1,2]. Other parameters affecting their structure are Spectroscopy (FTIR), X-ray Diffraction (XRD), Nuclear Magnetic
the curing process and the aging period [3,4]. Resonance (NMR) Spectroscopy and Thermogravimetric Analysis
Alkali activated materials have attracted considerable attention (TG) [12–16].
in the last three decades due to their properties that render them The alkali activation potential of industrial wastes such as fly
suitable as alternative binders in the construction industry, as fire/- ash from thermal power plants, slag from the steel industry and
corrosion resistant materials or as matrices for the encapsulation red mud from alumina production, has been extensively investi-
gated. Fly ash has been considered for the production of elements
⇑ Corresponding author. with suitable physico-chemical and mechanical properties that can
E-mail address: komni@mred.tuc.gr (K. Komnitsas).
be used in the construction industry or for immobilization of toxic

http://dx.doi.org/10.1016/j.conbuildmat.2016.06.051
0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
D. Zaharaki et al. / Construction and Building Materials 121 (2016) 686–693 687

elements [17–21]. High or low calcium slags have been alkali acti- include among others reduction of the consumption of natural
vated for the production of materials that can be used in various resources, minimization of the volume of wastes that will ulti-
industrial applications [22–27]. Other recent studies have shown mately be landfilled and reduction of the emission of greenhouse
that wastes such as fly ash, granulated blast furnace slag (GBFS) gases (GHGs) [44–46].
and calcium carbide residue can be alkali activated and used as The present experimental study aims to investigate the co-
binders to replace Ordinary Portland Cement (OPC) for the produc- valorization of C&D and industrial wastes, namely ferronickel slag
tion of ‘‘green” concrete as well as for the production of other con- and red mud, through alkali activation and define the main factors
struction elements with lower carbon footprint [28–31]. The alkali that affect the properties of the produced materials. The proposed
activation potential of red mud mixed with other industrial wastes approach is in line with the ‘‘zero-waste” and ‘‘circular economy”
has been also exploited for the production of bricks, tiles and fire principles of the European Commission and contributes to ‘‘closing
resistant construction materials [32,33]. the loop” of product lifecycles through greater recycling and re-
On the other hand, the alkali activation potential of C&D wastes use, while it brings maximum benefits for both the environment
still remains a challenge and a rather limited number of studies and the economy.
have been carried out so far. More specifically, Allahverdi and Kani
[34,35] produced geopolymer cement using waste bricks and con-
crete in different mixing ratios. They found that waste bricks 2. Experimental procedure
exhibited better alkali activation potential compared to concrete.
2.1. Materials
Ahmari et al. [36] produced geopolymeric binder from mixtures
of waste concrete and fly ash. When high alkaline NaOH solutions The raw materials used in the present study were:
were used (10 M) the compressive strength of the final products
(i) electric arc furnace slag from the ‘‘LARCO S.A” ferronickel plant in Larymna,
reached almost 30 MPa. Reig et al. [37] investigated the properties
prefecture of Lokris, central Greece,
and microstructure of alkali-activated cement pastes and mortars (ii) C&D waste components i.e. tiles, bricks and concrete collected from various
produced from red clay brick waste. They optimised the type and demolished buildings in the city of Chania, island of Crete, Greece and
concentration of alkali activator to produce mortars with compres- (iii) red mud from ‘‘Aluminium of Greece”, Agios Nikolaos plant, prefecture of
sive strength up to 50 MPa after 7 days of curing at 65 °C. Sun et al. Veotia, central Greece.

[38] produced geopolymer-type materials by activating waste


All materials were pulverized using a BICO Type UA pulveriser and a FRITSCH
ceramics. The synthesized products exhibited very good thermal
pulveriser and homogenized. Grain size analysis was determined using a Master-
stability and compressive strength of over 75 MPa. Pathak et al. sizer S (Malvern Instruments) particle size analyser (Table 1). The chemical compo-
[39] investigated two types of C&W wastes, i.e. concrete powder sition of the raw materials, as derived from an X-ray fluorescence energy dispersive
and brick powder, to produce geopolymers. Calorimetric analysis spectrometer (XRF–EDS) Bruker-AXS S2 Range type, is presented in Table 2. Loss on
ignition (LOI) was determined by heating each raw material at 1050 °C for 4 h.
showed that brick dust is more reactive than concrete waste and
the final products exhibited maximum compressive strength over
10 MPa. It was deduced from their study that the physico- 2.2. Methodology
mechanical properties of geopolymers obtained from these C&D
waste components were comparable with those of other building In the present study, raw materials were used after grinding in various mix-
materials. Zedan et al. [40] investigated the effect of mixing cera- tures. The % w/w of each raw material subjected to alkali activation as well as
the codes of all specimens are presented in Table 3. The percentage of each raw
mic, red clay brick and concrete wastes with ground granulated material was selected by taking into account its mineralogical composition and pre-
blast furnace slag (GGBFS) on the properties of the produced alkali vious studies carried out by the authors.
activated slag cement. It is mentioned that GGBFS has a much Raw materials were mixed under continuous stirring with the activating solu-
lower Fe2O3 and higher CaO content than the ferronickel slag used tion, which was prepared by dissolving NaOH anhydrous pellets (Sigma Aldrich)
in distilled water and then adding sodium silicate solution (Na2O = 7.5–8.5%,
in this study.
SiO2 = 25.5–28.5%, Merck). The final solution was allowed to cool at room temper-
All these studies prove that most components of C&D waste ature for 24 h prior to use.
entail a noticeable alkali activation potential. This potential may The required amount of NaOH pellets was dissolved in distilled water to obtain
be enhanced if specific ratios of components are mixed to adjust solution molarity 8, 10 or 12 M. The liquid/solid (L/S) ratio varied slightly between
the mineralogical composition of the starting raw materials prior 0.27 and 0.33 in order to improve the flowability characteristics of the produced
paste before casting. The fresh paste was cast in cubic metal moulds of 5 cm edge,
to alkali activation [41]. It has to be underlined that the European which were vibrated for a few minutes to eliminate the presence of air voids in the
Commission has identified C&D wastes as a priority waste flow for reactive mass and improve the properties of the final specimens. The moulds
reuse and highlighted the noticeable environmental benefits of remained at room temperature for 4 h to allow early initiation of reactions which
their valorization (Directive 2008/98/EC) [42,43]. These benefits favour the development of structural bonds and solidification of the paste. Then
the specimens were demoulded and sealed in plastic bags to prevent fast evapora-
tion of water during curing.
Table 1
The specimens were then heated at 80 °C in a laboratory oven (ON-02G) for
Particle size (lm) of raw materials.
24 h, allowed to cool and after aging at room temperature for 7 days were subjected
Slag Tiles Bricks Concrete Red mud to compressive strength testing using a MATEST C123N load frame. The thermal
behaviour of the specimens was evaluated after heating at high temperature
Size (lm) <120 <140 <140 <190 <76 (400, 600 and 800 °C) for 1 h in a laboratory furnace N-8L Selecta; weight loss, vol-
d50 (lm) 12 14 7 10 4 umetric shrinkage and porosity (using a Micromeritics AutoPore 9400 porosimeter)
were also recorded. All tests and measurements were carried out in triplicate.

Table 2
Chemical composition (%) of raw materials.

Material Fe2O3 SiO2 Al2O3 CaO Na2O K2O MgO TiO2 Cr2O3 LOI SUM
Slag 43.83 36.74 9.32 3.73 – – 2.76 – 2.82 – 99.2
Tiles 5.39 70.54 9.80 8.78 – 1.37 4.46 0.77 – – 101.1
Bricks 6.00 57.79 14.95 8.79 1.03 2.80 4.75 0.85 – 1.89 98.9
Concrete 0.75 5.81 1.49 65.42 0.57 1.26 4.21 0.03 – 21.59 101.1
Red mud 41.65 9.28 15.83 10.53 2.26 0.21 – 4.73 – 16.77 101.3
688 D. Zaharaki et al. / Construction and Building Materials 121 (2016) 686–693

Table 3 from raw material mixtures containing lower percentages of con-


% w/w of each raw material used for the production of each alkali activated specimen. crete, i.e. 10 and 15% w/w (specimens 50S-20T-20B-10C and 25S-
Code Slag Tiles Bricks Concrete Red mud 30T-30B-15C, respectively). When the percentage of concrete
S 100 – – – – increases to 30% w/w (specimen 50S-10T-10B-30C) the strength
T – 100 – – – drops to just below 60 MPa. When the molarity increases to
B – – 100 – – 12 M NaOH, the strength is slightly decreased compared to 10 M
C – – – 100 – NaOH, by almost 4%, for all alkali activated combinations tested.
R – – – – 100
50S-20T-20B-10C 50 20 20 10 –
This is maybe due to the presence of unreacted residual alkalis
25S-30T-30B-15C 25 30 30 15 – which makes solidification more slow and results in partial loss
50S-10T-10B-30C 50 10 10 30 – of strength. Lower molarity of NaOH (8 M) does not seem to pro-
90S-10R 90 – – – 10 vide the overall required alkalinity in the initial paste and thus less
80S-20R 80 – – – 20
Si and Al ions are dissolved from the raw materials and participate
70S-30R 70 – – – 30
60S-40R 60 – – – 40 in the formation of stronger bonds. It is thus deduced that the opti-
50S-50R 50 – – – 50 mum NaOH molarity for alkali activation of slag and C&D waste
components is 10 M.
XRD analysis of the raw materials and the produced specimens was performed Fig. 2 shows the compressive strength of slag-red mud speci-
using a Bruker AXS (D8 Advance type) diffractometer with Cu tube, scanning range mens for various mixing ratios vs. the molarity of NaOH. It is
from 4° to 70° 2h, step 0.02° and measuring time 0.2 s/step. The qualitative analysis shown from this figure that the optimum NaOH molarity is also
was assessed with the use of the DIFFRACplus EVA v. 2006 software and the Powder 10 M, as discussed in Fig. 1. The compressive strength of the pro-
Diffraction File (PDF-2) database. TG analysis was carried out using a TGA-6/DTG
Perkin Elmer differential thermogravimetric analyser (temperature measurement
duced specimens decreases gradually when increased red mud
precision ±2 °C, microbalance sensitivity <5 lg). The experiments were carried addition percentages are used in the initial mixtures and reaches
out at atmospheric pressure under nitrogen atmosphere, with a flow rate of almost 40 MPa when the slag:red mud ratio is 1:1 (50S-50R spec-
100 mL/min and a heating rate of 10 °C/min. FTIR analysis was carried out using imen), which is still an acceptable value for several construction
pellets produced after mixing a pulverized sample of each specimen with KBr at
applications. By taking into account that red mud alone cannot
a ratio of 1:100 w/w, using a Perkin-Elmer Spectrum 1000 spectrometer (USA).
SEM analysis was performed with a JEOL 6380LV scanning electron microscope be alkali activated, as indicated by the compressive strength of
equipped with an EDS INCA microanalysis system with low vacuum, pressure the control R specimen which is just 2.5 MPa, it is deduced that
30 Pa, voltage 20 kV and 10–12 mm working distance from the detector. Samples in order to obtain materials with adequate compressive strength
were mounted in resin and thin polished sections were prepared and used for that can be used in the construction industry, red mud should be
SEM studies.
mixed with other suitable raw materials that improve the proper-
ties of the initial mixture.
3. Results and discussion
3.2. Effect of the molar ratios of oxides present in the initial paste
3.1. Effect of activating solution molarity on compressive strength
Table 4 shows the molar ratios of various oxides present in the
Fig. 1 shows the compressive strength of the specimens pro- initial paste, aiming to elucidate the role of the mineralogy of the
duced by mixing ferronickel slag with C&D wastes in various per- raw materials during alkali activation. In this table the compressive
centages as a function of the molarity of the activating solution (8, strength of all specimens, including the control ones produced
10 or 12 M NaOH); specimen codes are presented in Table 3. It also from slag, single components of C&D waste and red mud during
shows the indicative compressive strength of the control speci- alkali activation with 10 M NaOH are also presented for
mens (S, T, B and C) produced at 10 M NaOH, for comparison. comparison.
It is seen from Fig. 1 that the strength of alkali activated speci- First of all, it is seen that since slag, tiles and to a lesser degree
mens is affected by the molarity of the activating solution. At 8 M bricks are successfully alkali activated, the compressive strength of
NaOH the strength of the specimens for all mixing combinations the specimens produced using noticeable percentages of these raw
tested is around 55 MPa, 30% lower than the strength of the control materials is also high. On the other hand, it is shown that higher
slag specimen S. When the molarity of the alkaline solution percentage of concrete and red mud in the initial paste results in
increases to 10 M NaOH a noticeable beneficial effect on the com- decrease of the compressive strength.
pressive strength for all specimens is shown. The highest values, It is known that specimens with high compressive strength are
around 76 MPa, are obtained for specimens which are produced usually obtained when high SiO2/Al2O3 molar ratios are present in
the initial paste [41]; these ratios are for example 7.33 and 12.67
90 80 °C, 24 h 8 M NaOH when only slag (S) or tiles (T) are alkali activated. High SiO2/Al2O3
S
10 M NaOH
80
Compressive strength (MPa)

12 M NaOH 90
80 oC, 24 h 8 M NaOH
70
Compressive strength (MPa)

T 80 10 M NaOH
60 70 12 M NaOH
50 B 60
40 50
30 40
20 30
10 C
20
0 10
0

Fig. 1. Compressive strength of specimens produced by mixing slag with C&D Fig. 2. Compressive strength of slag-red mud specimens as a function of NaOH
wastes (error bars denote standard deviation of three specimens). molarity (error bars denote standard deviation of three specimens).
D. Zaharaki et al. / Construction and Building Materials 121 (2016) 686–693 689

Table 4 and the reactivity of the most important phases present in raw
Molar ratios of oxides present in the initial paste (10 M NaOH) and compressive materials define the properties of the final products. It is known
strength of alkali activated products.
that the solubility of Si, Al, and Fe is very important [2,49]. It is also
Code SiO2
Al2 O3
SiO2
CaO
SiO2
ðAl2 O3 þCaOÞ
SiO2
ðAl2 O3 þFe2 O3 Þ
H2 O
ðNa2 OþK2 OÞ
Strength known that NaOH dissolves more Si and Al from raw materials
(MPa) compared to KOH [50]. The role of calcium is also important and
S 7.33 8.97 4.03 1.68 8.30 76.1 the formation of C-S-H and C-A-S-H phases is also possible if the
T 12.67 7.76 4.81 9.38 9.03 57.8 raw materials contain elevated calcium content, as in the case of
B 6.84 6.38 3.30 5.44 8.32 39.4
C 9.86 0.12 0.12 7.47 6.62 7.8
recycled concrete and red mud. On the other hand, it has to be
R 1.33 1.10 0.60 0.50 8.11 2.5 taken into account that the solubility of calcium decreases at ele-
50S-20T- 8.5 3.62 2.54 3.18 6.89 76.0 vated pH and results in the formation of unstable calcium hydrox-
20B-10C ide, which induces atmospheric carbonation and thus specimens
25S-30T- 8.86 2.95 2.21 4.54 6.10 74.1
with lower strength are obtained [51,52].
30B-15C
50S-10T- 8.26 1.38 1.18 2.67 6.96 59.2
10B-30C 3.3. Thermal stability
90S-10R 6.26 7.04 3.31 1.55 7.80 65
80S-20R 5.44 5.76 2.80 1.43 7.42 56
70S-30R 4.74 4.77 2.38 1.32 7.16 53.2 Fig. 3 presents the compressive strength of selected specimens
60S-40R 4.07 3.93 2.00 1.19 6.84 39.8 prepared from slag (S) as well as from mixtures of slag with C&D
50S-50R 3.53 7.20 1.70 1.08 6.66 38 waste (25S-30T-30B-15C) or red mud (90S-10R, 50S-50R) and then
subjected to high temperature heating at 400, 600 and 800 °C for
one hour. The compressive strength of the control specimens,
molar ratios facilitate the formation of dense and strong Si–O–Al which were not subjected to high temperature heating, is also
and Si–O–Si bonds. given for comparison.
When the SiO2/Al2O3 ratio is very low, such as the one present It is shown that for all specimens strength decreases gradually
in red mud paste (1.33), very few aluminosilicate bonds are formed when temperature increases from 400 to 800 °C. The best beha-
and thus the final strength is also low (2.5 MPa for R specimen), as viour is shown for specimens S, prepared only from slag, and
discussed in Section 3.4 and in other studies [47,48]. The very low 25S-30T-30B-15C, prepared by mixing slag with C&D waste com-
strength of red mud specimens is also related to the low values of ponents; despite the fact that their strength decreases by almost
the ratios SiO2/CaO and SiO2/(Al2O3 + CaO) as shown in Table 4 60% at 800 °C, it still remains at acceptable values for structural
(1.10 and 0.60, respectively), due to the low SiO2 content (9.28%) materials when subjected to thermal heating, close to 30 MPa.
of red mud. Also, the SiO2/(Al2O3 + Fe2O3) molar ratio is very low, Thus, these materials may be used for passive fire protection in
just 0.5 for the R paste. It is thus deduced that when both ratios constructions, as also reported by Sakkas et al. [53]. It is also shown
SiO2/Al2O3 and SiO2/(Al2O3 + Fe2O3) are low in the starting mixture in Fig. 3 that the slag-red mud specimen 90S-10R which was pro-
the final compressive strength of the specimens will be low. When duced from a starting mixture with higher slag content (90% w/w)
both these ratios increase the compressive strength also increases also retains an acceptable strength value after heating at 800 °C
as also shown for all combinations of mixtures between slag and (20 MPa) which is much higher than the strength of 50S-50R
red mud (90S-10R to 50S-50R). When the SiO2/(Al2O3 + Fe2O) ratio specimen (8 MPa), in which the slag to red mud weight ratio is
is low, specimens can only acquire high strength if the SiO2/Al2O3 1:1.
ratio is high, as shown for the S specimen. The decrease in the compressive strength of specimens that
Regarding concrete, it is deduced from Table 4 that SiO2/Al2O3 were subjected to high temperature heating is mainly related to
and SiO2/CaO are the critical ratios. Concrete has by far the lowest the increased weight loss and volumetric shrinkage that reach
SiO2 (5.81%) and Al2O3 (1.49%) and the highest CaO content, thus 11% and 8%, respectively. Also the decrease in porosity, which for
despite the fact that the SiO2/Al2O3 ratio is high in the initial paste example for specimen 50S-50R heated at 800 °C reaches 25%, is
(9.86) both SiO2/CaO and SiO2/(Al2O3 + CaO) ratios are extremely an important factor that affects the mechanical properties of the
low (0.12), and since no alkali activation occurs to dissolve raw produced specimens since it is accompanied by the development
materials and generate Si and Al ions that result in the formation of easily visible microcracks. The same behaviour has been also
of aluminosilicate bonds, the compressive strength of the produced observed in one of our recent studies [41]. All these factors cause
specimens is very low. deterioration of the structural integrity of the specimens as a result
On the other hand, as Table 4 data shows concrete can be mixed of the partial decomposition of Si–O–Al and Si–O–Si bonds, as also
with slag and other C&D components, up to 30% w/w in the start- discussed in other studies [54–58].
ing mixture, and alkali activated without affecting noticeably the
final strength of the produced specimens. This is due to the fact
that both critical ratios, as mentioned earlier, are improved (spec- 90 80 oC, 24 h S
imens 50S-20T-20B-10C, 25S-30T-30B-15C and 50S-10T-10B-30C).
Compressive strength, MPa

80 25S-30T-30B-15C
The ratio H2O/(Na2O + K2O) in the initial paste does not seem to 70 90S-10R
affect the final compressive strength since it is high in all combina- 50S-50R
60
tions of raw materials used. It only shows that adequate water was 50
available in the reactive paste allowing the initiation of alkali acti-
40
vation reactions. Whether or not these reactions will progress and
30
to which degree depends on the presence of appropriate percent-
20
ages of the required mineralogical phases so that optimum ratios
10
among SiO2, Al2O3 and CaO are present, as discussed earlier, under
the condition that sufficient concentration of alkali activator is 0
Control 400 600 800
used. oC
It has to be underlined though that apart from the above men-
tioned ratios, which indicate the alkali activation potential of Fig. 3. Compressive strength of selected specimens subjected to high temperature
specific raw materials, the strength of alkali activation solution heating (error bars denote standard deviation of three specimens).
690 D. Zaharaki et al. / Construction and Building Materials 121 (2016) 686–693

Fig. 4. XRD patterns of selected specimens produced from slag, C&D waste components and red mud (specimen codes are presented in Table 3).

100 Table 5
S
94 FTIR spectra band assignments.
T
Band number, Assignment
88 B cm1
Weight, %

82 25S-30T-30B-15C 3440, 2950, H–O–H stretching and bending vibrations


76 2340, 1830
1490, 1420, 880 Atmospheric carbonation, out of plane bending of CO3
70 1050 Si–O stretching vibrations and Si–O–Si or Al–O–Si
C asymmetric stretching vibrations
64 710–460 In plane Si–O bending and Al–O linkages as well as
58 bending Si–O–Si and O–Si–O vibrations
25 200 375 550 725
Temperature, o C
tain various crystalline phases such as quartz and calcite, while red
Fig. 5. TG curves of selected specimens produced from slag and C&D waste mud consists mainly of hematite, goethite and diaspore [2,41].
components (specimen codes are presented in Table 3). Fig. 4 presents the XRD patterns of three selected specimens
produced from mixtures of slag, C&D waste components and red
3.4. Morphology of final products mud as well as of the respective control specimens produced after
alkaline activation with the use of 10 M NaOH. It is seen from this
Our previous studies show that the mineralogy of ferronickel figure that all specimens consist of partially reacted or unreacted
slag is dominated by the presence of amorphous silicate or alumi- crystalline phases, such as quartz, calcite, hematite and albite,
nosilicate phases exceeding 50% w/w. C&D waste components con- which were also present in the raw materials. The formation of

C
710 1068
2950
B 800 1490

T 1420
2340 3440
460
1830
S
1050 1470

25S-30T-30B-15C 1440
880
R
50S-50R (600 oC)
50S-50R
400 900 1400 1900 2400 2900 3400
cm-1

Fig. 6. FTIR spectra of selected specimens produced from slag, C&D waste components and red mud (specimen codes are presented in Table 3).
D. Zaharaki et al. / Construction and Building Materials 121 (2016) 686–693 691

pirssonite, (Na2Ca(CO3)22H2O), in specimens T, C and B is due to such as B and T, and acquire also high strength, the peaks between
atmospheric carbonation. 27° and 42° 2-theta become broader. On the other hand, limited
It is also seen in Fig. 4 that slag-based specimens (S) which amorphicity is shown in the XRD patterns of C and R specimens
acquire the highest compressive strength are characterized by cer- which acquire low strength; in these specimens only unreacted
tain degree of amorphicity [59]. In specimens produced from C&D crystalline phases are present. It is also shown that no significant
components which also exhibit quite high degree of amorphicity, changes are present in the XRD pattern of 50S-50R specimen

Fig. 7. SEM images and EDS analysis of selected specimens produced from slag, C&D waste components and red mud (specimen codes are presented in Table 3).
692 D. Zaharaki et al. / Construction and Building Materials 121 (2016) 686–693

heated at 600 °C, compared to the respective non-heated 50S-50R illustrated in Fig. 7b and c, respectively. High content of Si and
specimen, apart from the fact that the crystallinity of some peaks Al, which are the main elements required for the formation of alu-
seems to be slightly reduced. minosilicate bonds, is detected in these specimens. In Fig. 7d an
Fig. 5 presents the TG curves, which denote the % weight loss indicative EDS spectrum is shown for specimen 25S-30T-30B-15C
versus temperature, for the control specimens produced from slag, (Si and Al are the main elements) for which the SEM image is pre-
bricks, tiles and concrete as well as for specimen 25S-30T-30B-15C sented in Fig. 7f.
produced after alkaline activation of a mixture comprising specific Finally, a certain degree of homogeneity is shown in 50S-50R
percentages of these raw materials with the use of 10 M NaOH. It is specimen (Fig. 7f) where dispersed angular grains of slag are visi-
shown from this figure that weight is gradually lost for S, T, B and ble. It is mentioned that the matrix of specimen R (Fig. 7g), that
25S-30T-30B-15C specimens. Up to 200 °C weight loss ranges is not well solidified and acquires very low strength (2.5 MPa) is
between 5 and 8% and is due to the loss of free water, while the quite different; various voids and pores are identified, while almost
total weight loss has also a narrow range, between 8 and 13%, no gel is formed due to limited alkali activation of red mud grains.
indicating that water is bound tightly in the hardened specimens. In this case Fe is the dominant element, while Si and Al are
Therefore, the noticeable strength of these specimens, ranging detected in very small quantities, as deduced from the EDS analysis
between 39 and 76 MPa, apart from the favourable mineralogical shown in Fig. 7h.
composition of the raw materials is also due to the appropriate
water content present in the initial paste that initiates favourable
4. Conclusions
reactions and results in the formation of a dense structure,
Mn((SiO2)z–AlO2)nwH2O.
The results of this study show that successful alkali activation of
Similar results (not shown) derived for slag-red mud specimens
C&D wastes, slag and red mud can be achieved at 80 °C, using
90S-10R, 80S-20R and 70S-30R, for which the total weight loss var-
NaOH as alkali activator, for the production of construction mate-
ies between 9 and 18% and the acquired strength is high, ranging
rials. More specifically, the final products acquire optimum proper-
from 53 to 65 MPa. On the other hand, the control concrete speci-
ties only if appropriate ratios of specific mineralogical constituents,
men (C) which acquires low compressive strength (7.8 MPa) shows
namely SiO2/Al2O3 and SiO2/CaO, are present in the starting mix-
entirely different behaviour. For this specimen, 9% of free water is
ture, provided of course that sufficient concentration of alkali acti-
lost at around 200 °C while the total weight loss, when the temper-
vator is used. In this case, alkali activation facilitates the formation
ature reaches 800 °C, increases to 41% due to the partial decompo-
of Si–O–Si and Al–O–Si bonds and the produced materials obtain
sition of calcite.
beneficial properties. Furthermore, it has been also shown that
Fig. 6 shows the FTIR spectra of selected specimens produced
final products obtained from specific starting mixtures exhibit
from slag, the components of C&D wastes and red mud, while
good thermal behaviour and can be used for fire protection.
the respective spectra band assignments are summarized in
Another important parameter during alkali activation is the pres-
Table 5. FTIR analysis confirms the presence of the major finger-
ence of sufficient water in the reactive paste which facilitates early
prints of the aluminosilicate matrix around 1050 cm1 which is
initiation of alkali activation reactions.
more obvious for high strength specimens, namely B, T, S and
The use of analytical techniques provides very useful insights
25S-30T-30B-15C. These bands are attributed to Si–O stretching
for the characterization and microstructure of the produced speci-
vibrations and Si–O–Si or Al–O–Si asymmetric stretching vibra-
mens. XRD analysis indicates the degree of amorphicity, as well as
tions and their presence indicates higher dissolution of the starting
the presence of partially reacted or unreacted crystalline phases in
materials, increased amorphicity and thus formation of specimens
all specimens. TG analysis shows the presence of tightly bound
with high compressive strength, as also discussed in other relevant
water in the matrix indicating if specimens will acquire a notice-
studies [15,60–62].
able strength. FTIR analysis confirms the presence of the major fin-
The band at 1050 cm1 is also shown, with a lower intensity, for
gerprints of the aluminosilicate alkali activated matrix. SEM
the 50S-50R specimen and the one heated at 600 °C for one hour,
analysis shows that the matrix of specimens with improved prop-
indicating that most of the aluminosilicate bonds are not broken
erties is more homogeneous as a result of the sufficient dissolution
during heating at this temperature. A slight shift though of the FTIR
of the raw materials used and the formation of aluminosilicate
spectra to the left implies that several chemical reactions have
bonds.
occurred. On the other hand, for the concrete specimen C that
acquires low strength only a small peak at around 1050 cm1 is
visible. Acknowledgements
The bands at 3440, 2950, 2340 and 1830 cm1 are assigned to
stretching and bending H–O–H vibrations of bound water mole- The authors would like to acknowledge (i) the financial support
cules. The bands between 1420 and 1490 cm1 as well as at of European Commission in the frame of Horizon 2020 project
880 cm1 are due to atmospheric carbonation and are attributed ‘‘Metal recovery from low-grade ores and wastes”, www.metgrow-
to the modes of CO3 in CaCO3. These bands are quite strong in C plus.eu, Grant Agreement n° 690088, and (ii) two anonymous
and 25S-30T-30B-15C specimens. The bands at around 460– reviewers for their constructive comments.
710 cm1 which are detected in all specimens are due to in plane
Si–O bending and Al–O linkages as well as Si–O–Si and O–Si–O References
bending vibrations [63,64].
[1] J. Davidovits, Geopolymers: inorganic polymeric new materials, J. Mater. Educ.
SEM images of selected specimens produced from slag, compo- 16 (1994) 91–139.
nents of C&D wastes and red mud are shown in Fig. 7. It is seen that [2] K. Komnitsas, D. Zaharaki, V. Perdikatsis, Effect of synthesis parameters on the
the matrix of specimens which acquire the highest compressive compressive strength of low-calcium ferronickel slag inorganic polymers, J.
Hazard. Mater. 161 (2009) 760–768.
strength is homogeneous due to the sufficient dissolution of the
[3] K. Somna, C. Jaturapitakkul, P. Kajitvichyanukul, P. Chindaprasirt, NaOH
raw materials and the formation of aluminosilicate gel (e.g. slag activated ground fly ash geopolymer cured at ambient temperature, Fuel 90
specimen S, Fig. 7a). (2011) 2118–2124.
The high strength acquired for the B and T specimens, that [4] M.J.A. Mijarsh, M.A. Megat Johari, Z.A. Ahmad, Synthesis of geopolymer from
large amounts of treated palm oil fuel ash: application of the Taguchi method
ranges between 40 and 58 MPa, is due to the development of net- in investigating the main parameters affecting compressive strength, Constr.
works which act as binders between the raw material grains, as Build. Mater. 52 (2014) 473–481.
D. Zaharaki et al. / Construction and Building Materials 121 (2016) 686–693 693

[5] F. Colangelo, R. Cioffi, F. Montagnaro, L. Santoro, Soluble salt removal from Labrincha, Y. Ding, J. de Brito (Eds.), Handbook of Recycled Concrete and
MSWI fly ash and its stabilization for safer disposal and recovery as road Demolition Waste, vol. 47, Woodhead Publishing Limited, Cambridge, UK,
basement material, Waste Manage. 32 (2012) 1179–1185. 2013, pp. 439–475.
[6] S. Hanjitsuwan, S. Hunpratub, P. Thongbai, S. Maensiri, V. Sata, P. [36] S. Ahmari, X. Ren, V. Toufigh, L. Zhang, Production of geopolymeric binder from
Chindaprasirt, Effects of NaOH concentrations on physical and electrical blended waste concrete powder and fly ash, Constr. Build. Mater. 35 (2012)
properties of high calcium fly ash geopolymer paste, Cem. Concr. Compos. 45 718–729.
(2014) 9–14. [37] L. Reig, M.M. Tashima, M.V. Borrachero, J. Monzó, C.R. Cheeseman, J. Payá,
[7] B. Singh, G. Ishwarya, M. Gupta, S.K. Bhattacharyya, Geopolymer concrete: a Properties and microstructure of alkali-activated red clay brick waste, Constr.
review of some recent developments, Constr. Build. Mater. 85 (2015) 78–90. Build. Mater. 43 (2013) 98–106.
[8] J. Yliniemi, J. Pesonen, M. Tiainen, M. Illikainen, Alkali activation of recovered [38] Z. Sun, H. Cui, H. An, D. Tao, Y. Xu, J. Zhai, Q. Li, Synthesis and thermal behavior
fuel–biofuel fly ash from fluidised-bed combustion: stabilisation/solidification of geopolymer-type material from waste ceramic, Constr. Build. Mater. 49
of heavy metals, Waste Manage. 43 (2015) 273–282. (2013) 281–287.
[9] S.A. Bernal, E.D. Rodríguez, A.P. Kirchheim, J.L. Provis, Management and [39] A. Pathak, S. Kumar, V.K. Jha, Development of building material from
valorisation of wastes through use in producing alkali-activated cement geopolymerization of construction and demolition waste (CDW), Trans.
materials, J. Chem. Technol. Biotechnol. (2016), http://dx.doi.org/10.1002/ Indian Ceram. Soc. 73 (2) (2014) 133–137.
jctb.4927. [40] S.R. Zedan, M.R. Mohamed, D.A. Ahmed, A.H. Mohammed, Effect of
[10] A. Nazer, J. Payá, M.V. Borrachero, J. Monzó, Use of ancient copper slags in demolition/construction wastes on the properties of alkali activated slag
Portland cement and alkali activated cement matrices, J. Environ. Manage. 167 cement, HBRC J., doi: 10.1016/j.hbrcj.2015.12.001.
(2016) 115–123. [41] K. Komnitsas, D. Zaharaki, A. Vlachou, G. Bartzas, M. Galetakis, Effect of
[11] E.I. Diaz, E.N. Allouche, S. Eklund, Factors affecting the suitability of fly ash as synthesis parameters on the quality of construction and demolition wastes
source material for geopolymers, Fuel 89 (2010) 992–996. (CDW) geopolymers, Adv. Powder Technol. 26 (2015) 368–376.
[12] K. Komnitsas, D. Zaharaki, Geopolymerisation: a review and prospects for the [42] Directive 2008/98/EC of the European Parliament and of the Council on waste
minerals industry, Miner. Eng. 20 (2007) 1261–1277. and repealing certain Directives, Official J. Eur. Union (2008) L.312/3.
[13] S. Ahmari, L. Zhang, Durability and leaching behavior of mine tailings-based [43] F. Pacheco-Torgal, Y. Ding, S. Miraldo, Z. Abdollahnejad, J.A. Labrincha, Are
geopolymer bricks, Constr. Build. Mater. 44 (2013) 743–750. geopolymers more suitable than Portland cement to produce high volume
[14] K.E. Hafid, M. Hajjaji, Effects of the experimental factors on the microstructure recycled aggregates HPC?, Constr Build. Mater. 36 (2012) 1048–1052.
and the properties of cured alkali-activated heated clay, Appl. Clay Sci. 116– [44] M. Bravo, J. De Brito, J. Pontes, L. Evangelista, Mechanical performance of
117 (2015) 202–210. concrete made with aggregates from construction and demolition waste
[15] D. Rieger, T. Kovářík, J. Říha, R. Medlín, P. Novotnỳ, P. Bĕlskỳ, J. Kadlec, P. Holba, recycling plants, J. Clean. Prod. 99 (2015) 59–74.
Effect of thermal treatment on reactivity and mechanical properties of alkali [45] H. Dahlbo, J. Bachér, K. Lähtinen, T. Jouttijärvi, P. Suoheimo, T. Mattila, S.
activated shale-slag binder, Constr. Build. Mater. 83 (2015) 26–33. Sironen, T. Myllymaa, K. Saramäki, Construction and demolition waste
[16] H. Zhang, V. Kodur, B. Wu, L. Cao, S. Qi, Comparative thermal and mechanical management – a holistic evaluation of environmental performance, J. Clean.
performance of geopolymers derived from metakaolin and fly ash, J. Mater. Prod. 107 (2015) 333–341.
Civ. Eng. 28 (2) (2016). doi: 10.1061/(ASCE)MT.1943-5533.0001359. [46] R.V. Silva, J. de Brito, R.K. Dhir, Properties and composition of recycled
[17] J.C. Swanepoel, C.A. Strydom, Utilisation of fly ash in a geopolymeric material, aggregates from construction and demolition waste suitable for concrete
Appl. Geochem. 17 (8) (2002) 1143–1148. production, Constr. Build. Mater. 65 (2014) 201–217.
[18] T. Bakharev, Geopolymeric materials prepared using Class F fly ash and [47] P. Duxson, J.L. Provis, G.C. Lukey, S.W. Mallicoat, W.M. Kriven, J.S.J. Van
elevated temperature curing, Cem. Concr. Res. 35 (2005) 1224–1232. Deventer, Understanding the relationship between geopolymer composition,
[19] J.Z. Xu, Y.L. Zhou, Q. Chang, H.Q. Qu, Study on the factors of affecting the microstructure and mechanical properties, Colloids Surf. A 269 (2005) 47–58.
immobilization of heavy metals in fly ash-based geopolymers, Mater. Lett. 60 [48] H.Y. Leong, D.E.L. Ong, J.G. Sanjayan, A. Nazari, The effect of different Na2O and
(6) (2006) 820–822. K2O ratios of alkali activator on compressive strength of fly ash based-
[20] Z. Zhang, H. Wang, Y. Zhu, A. Reid, J.L. Provis, F. Bullen, Using fly ash to partially geopolymer, Constr. Build. Mater. 106 (2016) 500–511.
substitute metakaolin in geopolymer synthesis, Appl. Clay Sci. 88–89 (2014) [49] J.G. Rapazote, C. Laginhas, A. Teixeira-Pinto, Development of building materials
194–201. through alkaline activation of construction and demolition waste (CDW) –
[21] F.U.A. Shaikh, V. Vimonsatit, Compressive strength of fly-ash-based resistance to acid attack, Adv. Sci. Technol. 69 (2010) 156–163.
geopolymer concrete at elevated temperatures, Fire Mater. 39 (2) (2015) [50] H. Xu, J.S.J. Van Deventer, G.C. Lukey, Effect of alkali metals on the preferential
174–188. geopolymerization of stilbite/kaolinite mixtures, Ind. Eng. Chem. Res. 40
[22] T.W. Cheng, J.P. Chiu, Fire-resistant geopolymer produced by granulated blast (2001) 3749–3756.
furnace slag, Miner. Eng. 16 (2003) 205–210. [51] K. Komnitsas, D. Zaharaki, V. Perdikatsis, Geopolymerisation of low calcium
[23] Z. Yunsheng, S. Wei, C. Qianli, C. Lin, Synthesis and heavy metal ferronickel slags, J. Mater. Sci. 42 (2007) 3073–3082.
immobilization behaviors of slag based geopolymers, J. Hazard. Mater. 143 [52] P.V. Krivenko, G.Yu. Kovalchuk, Directed synthesis of alkaline aluminosilicate
(1–2) (2007) 206–213. minerals in a geocement matrix, J. Mater. Sci. 42 (2007) 2944–2952.
[24] K. Komnitsas, D. Zaharaki, G. Bartzas, Effect of sulphate and nitrate anions on [53] K. Sakkas, P. Nomikos, A. Sofianos, D. Panias, Sodium-based fire resistant
heavy metal immobilisation in ferronickel slag geopolymers, Appl. Clay Sci. 73 geopolymer for passive fire protection, Fire Mater. 39 (3) (2015) 259–270.
(2013) 103–109. [54] Z. Pan, J.G. Sanjayan, B.V. Rangan, An investigation of the mechanisms for
[25] Y. Pontikes, L. Machiels, S. Onisei, L. Pandelaers, D. Geysen, P.T. Jones, B. strength gain or loss of geopolymer mortar after exposure to elevated
Blanpain, Slags with a high Al and Fe content as precursors for inorganic temperature, J. Mater. Sci. 44 (7) (2009) 1873–1880.
polymers, Appl. Clay Sci. 73 (2013) 93–102. [55] M. Guerrieri, J.G. Sanjayan, Behavior of combined fly ash/slag-based
[26] M. Albitar, M.S. Mohamed Ali, P. Visintin, M. Drechsler, Effect of granulated geopolymers when exposed to high temperatures, Fire Mater. 34 (4) (2010)
lead smelter slag on strength of fly ash-based geopolymer concrete, Constr. 163–175.
Build. Mater. 83 (2015) 128–135. [56] P.N. Lemougna, K.J.D. MacKenzie, Synthesis and thermal properties of
[27] M. Salman, Ö. Cizer, Y. Pontikes, R. Snellings, L. Vandewalle, B. Blanpain, K.V. inorganic polymers (geopolymers) for structural and refractory applications
Balen, Cementitious binders from activated stainless steel refining slag and the from volcanic ash, Ceram. Int. 37 (8) (2011) 3011–3018.
effect of alkali solutions, J. Hazard. Mater. 286 (2015) 211–219. [57] S. Chuah, W.H. Duana, Z. Pan, E. Hunter, A.H. Korayem, X.L. Zhao, F. Collins, J.G.
[28] J.L. Provis, P. Duxson, J.S.J. Van Deventer, The role of particle technology in Sanjayan, The properties of fly ash based geopolymer mortars made with dune
developing sustainable construction materials, Adv. Powder Technol. 21 sand, Mater. Des. 92 (2016) 571–578.
(2007) 2–7. [58] W. Ren, J. Xu, E. Bai, Strength and ultrasonic characteristics of alkali-activated
[29] G. Habert, J.B. D’Espinose De Lacaillerie, N. Roussel, An environmental fly ash-slag geopolymer concrete after exposure to elevated temperatures, J.
evaluation of geopolymer based concrete production: reviewing current Mater. Civ. Eng. 28 (2) (2016). doi: 10.1061/(ASCE)MT.1943-5533.0001406.
research trends, J. Clean. Prod. 19 (2011) 1229–1238. [59] D. Zaharaki, K. Komnitsas, V. Perdikatsis, Use of analytical techniques for
[30] K. Komnitsas, Potential of geopolymer technology towards green buildings and identification of inorganic polymer gel composition, J. Mater. Sci. 45 (2010)
sustainable cities, Procedia Eng. 21 (2011) 1023–1032. 2715–2724.
[31] A. Arulrajah, A. Mohammadinia, I. Phummiphan, S. Horpibulsuk, W. [60] W.K.W. Lee, J.S.J. Van Deventer, The effects of inorganic salt contamination on
Samingthong, Stabilization of recycled demolition aggregates by the strength and durability of geopolymers, Colloid Surf. A 211 (2–3) (2002)
geopolymers comprising calcium carbide residue, fly ash and slag 115–126.
precursors, Constr. Build. Mater. 114 (2016) 864–873. [61] J.G.S. Van Jaarsveld, J.S.J. Van Deventer, G.C. Lukey, The effect of composition
[32] W. Hajjaji, S. Andrejkovičová, C. Zanelli, M. Alshaaer, M. Dondi, J.A. Labrincha, and temperature on the properties of fly ash- and kaolinite-based
F. Rocha, Composition and technological properties of geopolymers based on geopolymers, Chem. Eng. J. 89 (2002) 63–73.
metakaolin and red mud, Mater. Des. 52 (2013) 648–654. [62] H.M. Khater, H.A. Abd el Gawaad, Characterization of alkali activated
[33] M. Zhang, T. El-Korchi, G. Zhang, J. Liang, M. Tao, Synthesis factors affecting geopolymer mortar doped with MWCNT, Constr. Build. Mater. 102 (2016)
mechanical properties, microstructure, and chemical composition of red mud- 329–337.
fly ash based geopolymers, Fuel 134 (2014) 315–325. [63] G. Socrates, Infrared and Raman Characteristic Group Frequencies, 3rd ed.,
[34] A. Allahverdi, E.N. Kani, Construction wastes as raw materials for geopolymer John Wiley & Sons Ltd, England, 2001.
binders, Int. J. Civ. Eng. 7 (3) (2009) 154–160. [64] A. Fernández-Jiménez, A. Palomo, Composition and microstructure of alkali
[35] A. Allahverdi, E.N. Kani, Use of construction and demolition waste (CDW) for activated fly ash binder: effect of the activator, Cem. Concr. Res. 35 (2005)
alkali-activated or geopolymer cements, in: E. Pacheco-Torgal, V.W.Y. Tam, J. 1984–1992.

You might also like