Materials Chemistry and Physics 199 (2017) 280e288

Contents lists available at ScienceDirect

Materials Chemistry and Physics

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

The influence of gibbsite in kaolin and the formation of berlinite on

the properties of metakaolin-phosphate-based geopolymer cements

a, b, *, Claus H. Rüscher b, E. Kamseu c, d, Jean N.Y. Djobo c, C. Leonelli d

K. Tchakoute
Herve
a
Laboratory of Applied Inorganic Chemistry, University of Yaounde I, Faculty of Science, Department of Inorganic Chemistry, PO. Box 812, Yaounde,
Cameroon
b €t Hannover, Callinstrasse 3, D-30167, Hannover, Germany
Institut für Mineralogie, Leibniz Universita
c
Local Materials Promotion Authority, PO Box: 2396, Nkolbikok, Yaound
e, Cameroon
d
Department of Engineering Enzo Ferrari, University of Modena and Reggio Emilia, Via Vivarelli 10, 41125, Modena, Italy

h i g h l i g h t s

Phosphoric acid solution were used as hardener for producing geopolymer cements.

Two kaolins containing about 11 and 28% of gibbsite were applied to produce geopolymer cement.

Geopolymers indicate the formation of berlinite which reinforced the geopolymers network.

The higher content of gibbsite in Kao2 affects negatively the properties of geopolymer cements.

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

Article history: Two different kaolins (Kao1, Kao2) containing about 11 and 28% of gibbsite were applied for producing
Received 17 February 2017 metakaolin-phosphate-based geopolymer cements. These kaolins were transformed to metakaolins by
Received in revised form calcination them at 700  C. Gibbsite contained in these kaolins were transformed to g-Al2O3 during the
14 June 2017
calcination. The hardener used in this work was a phosphoric acid solution with molar concentration
Accepted 5 July 2017
Available online 6 July 2017
10 M. It could be observed that the compressive strength of metakaolin-phosphate-based geopolymer
cements (54.41/36.01 MPa) decreases in the course GMK1/GMK2. The both geopolymer cements indicate
the formation of berlinite which is well dispersed in GMK1 and therefore reinforced the structure of the
Keywords:
Metakaolin
specimen. The micrograph images of phosphate-based geopolymer cements GMK2 is heterogeneous
Gibbsite while the one of the GMK1 is homogeneous and more compact microstructure. The higher content of
Hardener gibbsite in Kao2 affects negatively the mechanical and microstructural properties of metakaolin-
Geopolymer cements phosphate-based geopolymer cements.
Berlinite © 2017 Elsevier B.V. All rights reserved.

1. Introduction geopolymerization reaction through an alkaline medium has been

Geopolymer is essentially a chemical compound consisting of
repeating units of silico-oxide (Si-O-Si), silico-aluminate (Si-O-Al-
O-), ferro-silico-aluminate (-Fe-O-Si-O-Al-O-), or alumino-
phosphate (-Al-O-P-O-), created through a process of geo-
polymerization [1]. The reaction takes place in alkaline (Naþ, Liþ,
Kþ, Ca2þ or acidic (phosphoric or humic acid) medium. The
* Corresponding author. Laboratory of Applied Inorganic Chemistry, University of
Yaounde I, Faculty of Science, Department of Inorganic Chemistry, PO. Box 812,
Yaounde, Cameroon.
E-mail addresses: htchak@yahoo.fr, hervetchakoute@gmail.com, htchakoute@

).
uy1.uninet.cm (H.K. Tchakoute
studied extensively, and the effect of the mineralogical composition
of a raw aluminosilicate on geopolymerization and final properties
have already been reported. The acidic medium for geopolymer
synthesis is still less investigated and there is scanty information on
the effect of some minerals contents in the raw aluminosilicate on
the final properties of resulting geopolymer cements. For example,
the role of illite, halloysite, gibbsite and quartz contents in kaolin
for geopolymer synthesis in the alkaline medium has been reported
by many authors [2e5]. It is worth recalling that the mineralogical
composition of an aluminosilicate is one of the most important
factors affecting geopolymerization and the final properties of
geopolymer cements. Because of the high aluminum oxide contents
(i.e. Al-rich minerals in some metakaolin) has been reported that

http://dx.doi.org/10.1016/j.matchemphys.2017.07.020
0254-0584/© 2017 Elsevier B.V. All rights reserved.
 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288 281

berlinite (AlPO4) could be formed as main phosphate products after Table 1

geopolymerization into phosphoric acid [6]. Berlinite is known to The chemical composition of the kaolins (Kao1 and Kao2) in mass percent.

be isostructural to quartz, and both have very similar physical Oxide Kao1 Kao2
properties. Due to the positive effect played by quartz on strength SiO2 47.2 39.09
development of geopolymer cement, it could be very interesting to Al2O3 35.1 39.44
investigate the same for berlinite. So, study the influence of gibbsite Fe2O3 0.46 0.74
in kaolin and the formation of berlinite on the properties of K2O 0.46 0.30
TiO2 0.49 1.14
metakaolin-phosphate-based geopolymer cements could be of
MgO 0.19 0.13
great interest. Na2O <0.1 /
The aim of this work was to investigate the influence of gibbsite CaO <0.1 /
contains in kaolin and the formation of berlinite on the properties SO3 <0.01 /
P2O5 / 0.06
of metakaolin-phosphate-based geopolymer cements. The influ-
Cr2O3 / /
ence of gibbsite contains in kaolin and the formation of berlinite MnO / /
cements have been studied using compressive strength, X-ray LOI 14.94 18.48
diffractometry (XRD), infrared spectroscopy and scanning electron
LOI: Loss on ignition at 1000  C.
microscope (SEM) coupled with energy dispersive X-ray spectros-
copy (EDX).
the removal of absorbing water and the second endothermic peak
is attributed to the dehydration effect of gibbsite in Kao1 and Kao2.
2. Experimental
The third event corresponds to the dehydroxylation of kaolinite
leading to the formation of metakaolin. The weak peak observes at
2.1. Materials
around 580  C in the DSC curves of Kao1 and Kao2 is associated
Local clays (Kao1, Kao2) from Cameroon were used as raw ma- with the polymorphic transformation of a-quartz to b-quartz.
terials for producing metakaolin-phosphate-based geopolymer Commercial phosphoric acid (H3PO4 85%, puriss. p.a., Reag. ACS,
cements. The both clays were crushed and sieve to 90 mm followed Reag. ISO, Reag. Ph. Eur.) was used to prepare acid reactive ingre-
by calcining (Nabertherm, Mod. LH 60/14) for 4 h at 700  C (heat- dient or acid hardener. The commercial available quartz was also
ing/cooling rate of 1 /min) to obtain metakaolins MK1 and MK2. used as reference material.
The obtained aforementioned metakaolins correspond to the The hardener or reactive ingredient was prepared by diluting
metakaolins from Kao1 and Kao2, respectively. The kaolins and commercial phosphoric acid in distilled water in order to get the
metakaolins were already used by Tchakoute
et al. [5] for producing reactive ingredient with molar concentration 10 M. The molar
metakaolin-based geopolymer cements using sodium waterglass as concentration 10 M has been chosen according to the findings work
of Tchakoute
et al. [7]. The obtained hardener was allowed to
a hardener. They reported that Kao1 and Kao2 content 80.9% and
64.5% of kaolinite, and 11.3% and 28.3% of gibbsite, respectively. mature in the laboratory at room temperature for 4 h prior to use in
Fig. 1 displays the X-ray patterns of Kao1 and Kao2. The quartz order to dilute completely the commercial phosphoric acid, and
content in MK1 and MK2 to be estimated around 10% and 8%, produce H2PO 4 in the hardener which is necessary for the second

respectively and the both metakaolins contained ɣ-Al2O3 and c-
Al2O3 [5]. The chemical compositions of both kaolins are given in
Table 1. Fig. 2 shows the thermal history of the DSC curves of Kao1
and Kao2. The DSC curves of Kao1 and Kao2 show four endothermic
peaks at around 50, 300 and 520 and 580  C. The first is ascribed to
step of geopolymerization.
Phosphate-based geopolymer cement pastes were prepared by
adding hardener gradually to each metakaolin and mechanically
mixing for 5 min, obtaining the samples GMK1 and GMK2. GMK1
and GMK2 are geopolymer cements from MK1 and MK2, respec-
tively. The liquid/solid mass ratio was kept constant at 0.95
obtaining suitable workability. The fresh geopolymer cements were
molded into a cylindrical PE-containers (20 mm in diameter and
40 mm in height), which were closed in order to hinder the
evaporation of water during the setting. The specimens were kept
at room temperature for five days and then cured at 60  C in an
oven for 24 h in order to accelerate the polycondensation process.
The obtained phosphate-based geopolymer cements were
demoulded and conserved at room temperature for 28 days before
compressive strength testing.

2.2. Characterizations

The influence of gibbsite in kaolin and the formation of berlinite

Fig. 1. X-ray patterns of kaolins (Kao1, Kao2).
(AlPO4) during the geopolymerization process on the mechanical
properties of phosphate-based geopolymer cements were evalu-
ated by measuring the compressive strength. It was measured us-
ing a manually driven testing machine (ENERPAC P392, USA).
Eleven samples were tested for each composition and the average
compressive strength values reported as the results. After
compressive strength measurement, the fragments of each spec-
imen were collected. One part of these fragments was used to
observe the morphological changes. The other parts were crushed
and the obtained powders were used to measure X-ray
282 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288
Fig. 2. DSC curves of Kao1 and Kao2.
diffractometry, thermal analysis (TG/DSC) and infrared spectros-
copy. XRD patterns were taken using CuKa radiation between 5 and
80 in 7 h in steps of 0.03 (Bruker D4). Infrared (IR) absorption
spectra were recorded by the KBr method (200 mg KBr, 1 mg
sample, Bruker Vertex 80v, 2 cm1, 32 scans). TG and DSC of
crushed geopolymer cements were performed using alumina cru-
cible between 25 and 1000  C (technical air with flow 20 ml/min,
10  C/min heating/cooling rate, Setaram Setsys Evolution 1650).
Pieces from the mechanical testing, after gold coating and drying,
were used for microstructure observations using a JEOL JSM-6390A
Scanning Electron Microscope (SEM) coupled to microanalysis by
energy-dispersive X-ray analysis (EDX) with an acceleration voltage
of 30.0 kV.
3. Results and discussion
3.1. Characterization of metakaolins (MK1 and MK2)
Fig. 3 illustrates the X-ray patterns of MK1 and MK2 and in-
dicates the presence of the broad hump structure between 14 and
32 (2q) with centered around 22 (2q). The intensity of this broad
hump structure is higher in the X-ray pattern of MK1 compared to
the one of the MK2. This suggests that MK1 is super-reactive than
MK2. Besides the presence of this halo diffraction, we observed the
crystalline peaks of illite, quartz, and anatase as impurities. We also
observed ɣ-Al2O3 at 67.1 (2q) which is resulting in the calcination
product of gibbsite. The intensity of this peak is higher in the XRD
pattern of MK2 suggesting the higher content of this component in
MK2.
The IR spectra of MK1 and MK2 are shown in Fig. 4. The ab-
sorption band that appears at 458 cm1 is assigned to the bending
vibration of Si-O. The bands at 551 and 559 cm1 in the IR spectra of
MK1 and MK2, respectively are ascribed to the sialate (Si-O-Al)
bonds with Al in 6-fold coordination. These bands could be
attributed to illite and ɣ-Al2O3. The absorption band at 693 cm1 is
assigned to Si-O of quartz. The band at 800 cm1 in the IR spectra of
MK1 and MK2 are ascribed to the amorphous silica. The intensity of
this band is higher in the IR spectrum of MK1 suggesting that MK1
content more amorphous phase. This result is consistent with the
XRD results which show the higher intensity of the broad hump
structure between 14 and 32 (2q) (Fig. 4). The bands at 1077 and
1079 cm1 in the IR spectra of MK1 and MK2, respectively are
 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288
Fig. 3. X-ray patterns of metakaolins (MK1, MK2).
attributed to Si-O-Al and Si-O-Si in metakaolins. The value of this
wavenumber is slightly lower in the IR spectrum of MK1 indicating
more Al includes in the tetrahedral structure of MK1 compared to
the one of the MK2. This entails the higher reactivity of MK1
compared to the one of the MK2. The bands at 1650 and 3536 cm1
were recorded in the IR spectra of both metakaolins. These bands
are assigned to the OH of water molecules.
3.2. Characterization of metakaolin-phosphate-based geopolymer
cements
3.2.1. X-ray patterns
The X-ray patterns of GMK1 and GMK2 are given in Fig. 5. These
X-ray patterns display the typical broad hump structure in the
range 16e38 (2q) for both specimens. This broad hump appears
Fig. 4. IR spectra of metakaolins (MK1, MK2).
283
Fig. 5. X-ray patterns of phosphate-based geopolymer cements.
between 14 and 32 (2q) in the XRD patterns of both metakaolins,
as can be seen in Fig. 3. The shift of this broad peak toward higher
2q in the both X-ray patterns of geopolymer cements induces the
formation of 3D poly(phospho-siloxane) eSi-O-P-O-Si- networks
due to the depolymerization of metakaolins by phosphoric acid
solution as a hardener. The additional minerals such as quartz and
anatase are also observed in the XRD patterns of both geopolymer
cements and metakaolins suggesting thus these minerals remain
unchanged during the geopolymerization. The absence of the peaks
of illite and g-Al2O3 could be related to the dealumination of
metakaolin Al-O layer and g-Al2O3 in acidic medium. The Al3þ
species contained in illite and g-Al2O3 are leached out, leaving re-
sidual silicate layer. The Al3þ species leached out during the second
step of geopolymerization react with PO3 4 from hardener leading
to the formation of berlinite which is isostructural with quartz.
Some peaks of this new component are superimposed with those of
quartz except those observed at 3.98 and 3.61 Å as (101) and (003),
respectively with low intensity (JCPDS No. 10-0423) in the XRD
patterns of both geopolymer cements (Fig. 5). In order to show the
formation of berlinite, the peaks of quartz were tentatively sub-
tracted of MK1 and MK2 (Fig. 6) and subtracted again in the XRD
patterns of both geopolymer cements (Fig. 7). The difference
spectra (Fig. 7) now show practically the peaks of berlinite. This
confirms the presence of berlinite in the structure of metakaolin-
phosphate-based geopolymer cements which could act as filler
and reinforced the structure of geopolymer cements.
3.2.2. Infrared spectroscopy
The infrared spectra of the metakaolin-phosphate-based geo-
polymer cements are shown in Fig. 8. The broad bands that appear
at about 1647 and 3438-3412 cm1 correspond to the stretching
and deformation vibration modes of H-O-H and O-H, respectively.
The absorption bands at 467 and 466 cm1 are attributed to S-O-Si
in-plane vibration. The one at 559-551 cm1 in the infrared spectra
of metakaolins is ascribed to Si-O-AlVI shifted toward a lower
wavenumber after geopolymerization at about 525 cm1. This
displacement could be related to the partial replacement of SiO4 by
284 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288
Fig. 6. Difference X-ray patterns of metakaolins and quartz substracted.
PO4 units, resulting in the change in the local chemical environ-
ment [8]. The absorption band that appears at 592 cm1 in the
infrared spectra of GMK1 and GMK2 and absent in those of meta-
kaolins (Fig. 3) can be ascribed to a dominant contribution from
AlO6 vibrations [9]. This band could be attributed to the presence of
berlinite, AlPO4, as discussed above. The presence of this new
compound is in accordance with those obtained by X-ray patterns
of phosphate-based geopolymer cements. The absorption band at
666 cm1 could be ascribed to the stretching vibration modes of Si-
O-Si bonds and the one at 805-800 cm1 could be related to Si-O-P
bands [10]. The intensity of this band is higher in the infrared
Fig. 7. Difference X-ray patterns of phosphate-based geopolymer cements and quartz
substracted.
Fig. 8. IR spectra of phosphate-based geopolymer cements.
spectrum of GMK1 compared to the one of the GMK2. This suggests
the formation of the long chain of -Si-O-P-O-Si- in the networks
leading to the higher densification of GMK1, as discussed below.
The bands appearing at 937 and 931 cm1 in the IR spectra of
geopolymer cements are attributed to P-O vibration modes in the
structure of geopolymer cements [8,10,11]. The absorption bands
observed at 1077 and 1079 cm1 (called density of states peaks
maximum, DOSPM, in the following) in the IR spectra of MK1 and
MK2, respectively shifted toward higher wavenumber at 1089 and
1094 cm1 for GMK1 and GMK2, respectively (Figs. 4 and 8). The
displacement of this DOSPM implies the change of the micro-
structure due to the depolymerization and polycondensation pro-
cess suggesting the formation of a long chain -Si-O-P-O-Si- of
poly(phospho-siloxo) which becomes stronger. The higher value
of the wavenumber of the main band of GMK2 could be related to
the more inclusion of PO4 unit in the network. The excess PO4 unit
in GMK2 is confirmed by the appearance of a more shoulder band
at about 1157 cm1 compared to the one that appears in the IR
spectrum of GMK1. The excess of PO4 units in GMK2 could be
related to the lower amorphous phase contained in MK2 because
during the preparation of both geopolymer cements we used the
same mass ratio hardener/MK. The excess of PO4 units in GMK2
could lead to the formation of PO4 units and make weaker the
structure of the specimen because of the presence of unbalanced
charges of (PO4)3, therefore affect negatively the compressive
strength of GMK2, as discussed below.
3.2.3. TG and DSC analysis
Figs. 9 and 10 display the TG and DSC curves of hardened
metakaolin-phosphate-based geopolymer cements, GMK1 and
GMK2 have been measured at the temperature range 25e1000  C.
The TG curves indicate that the total mass loss is about 21.77 and
20.99% for GMK1 and GMK2, respectively. It can be seen that the TG
and DSC curves of geopolymer cements contain three types of
water that will escape during the heating such as physically bonded
water, chemically bonded water and hydroxyl groups [12]. TG
curves of the samples in Fig. 9 records the absolute mass loss. We
observed that the samples go through three distinct phases viz the
first mass loss occurs from 25 to 100  C are due to the physically
bonded water which is 0.86 and 0.65% for GMK1 and GMK2,
 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288
Fig. 9. TG of phosphate-based geopolymer cements.
Fig. 10. DSC of phosphate-based geopolymer cements.
respectively. The second occurs between 100 and 300  C accom-
panied by a broad endothermic phenomenon in the DSC curves
(Fig. 10) is associated with the chemically bonded water and
correspond to the loss of 19.57 and 18.85% of water for GMK1 and
GMK2, respectively. The third mass loss appears at the temperature
above 300  C due to the dehydroxylation of OH groups and corre-
sponds to the loss of 1.23 and 1.15% of water for GMK1 and GMK2,
respectively. The lower endothermic peaks around 270  C in the
DSC curves of GMK1 and GMK2 but not clearly observed in the TG
curves indicate that monoaluminum phosphate undergoes trans-
formation to amorphous phase [13e19] and also small amounts of
some hydrated aluminum phosphates, can be formed
[13e15,18e20]. This result confirms the formation of berlinite
during the geopolymerization process and corroborates with the
results of XRD (Fig. 6) and IR spectroscopy (Fig. 8). A weak endo-
thermic peak observed at around 584  C in the DSC curves for
GMK1 and GMK2 is ascribed to the polymorphic transformation of
a-berlinite to b-berlinite [21,22]. This peak is clearly observed in the
DSC curve of GMK2 suggesting thus the formation of more berlinite
in GMK2 due to the presence of more Al2O3 contained in the raw
material (Kao1), as can be seen in Table 1. a-b quartz inversion has
been fixed generally at 573  C. Therefore, it can be surmised that
more aluminum oxide is present in the raw materials, more is the
amount of berlinite formed.
3.2.4. Compressive strength
The compressive strength of metakaolin-phosphate-based
geopolymer cements GMK1 and GMK2 measured after 28 days in
285
the atmosphere of the laboratory are 54.41 and 36.01 MPa,
respectively. The higher compressive strength of geopolymer ce-
ments from phosphoric acid as a hardener could be related firstly to
the more connectivity of poly(phospho-siloxo) eSi-O-Si-O-P-O-
networks and secondly to the formation of berlinite which
dispersed in the matrix and could act as filler and therefore rein-
forced the structure of geopolymers [8,23]. The higher connectivity
of GMK1 is a good agreement of the IR results which indicate that
the higher intensity of the band at 800 cm1. It is important to note
that Tchakoute
and Rüscher [23] used kaolin which does not
contain gibbsite for producing phosphate-based geopolymer ce-
ments and the compressive strength value of metakaolin-
phosphate-based geopolymer cement was 93.8 MPa. The chemi-
cal composition of Kao1 and Kao2 was given in Table 1 and in-
dicates that the aforementioned raw materials Kao1 and Kao2
content 35.10 and 39.44% of Al2O3, respectively and 47.20 and
39.09% of SiO2, respectively. The higher amount of Al2O3 in Kao2
can also be related to its high content of gibbsite. It is obvious that
GMK2 contains a higher amount of berlinite compared to the
GMK1, but the higher minerals content of kaolinite in Kao1 entail
the formation of a higher amount of amorphous phase in MK1 and
its high reactivity for geopolymerization. It was typically found that
the higher SiO2 contained in MK1 could be responsible for the
formation of a long chain of siloxo in the matrix of GMK1. The lower
compressive of the specimen GMK2 could be related to the lower
amorphous phase (metakaolinite) content in MK2. The lower
amorphous phase in MK2 could be related to the higher amount of
gibbsite in Kao2.
3.2.5. SEM images
The scanning electron micrograph images and EDS analysis of
geopolymer cements GMK1 and GMK2 are given in Fig. 11AeE. The
micrograph images of phosphate-based geopolymer cements
GMK2 (Fig. 11, B) is heterogeneous. From the microstructure
observation and EDS analysis, there are three phases: a light dark
particulate matrix of high silicon concentration (Fig. 11, F), a white
matrix constitute with Si-O-Si-O-P-O (Fig. 11, E) and the black
matrix (Fig. 11, D) with slightly high phosphate concentration. The
black color of GMK2 could be related to the presence of an excess of
PO4 units in the matrix which weakens the structure of the spec-
imen GMK2 and entails the lower compressive strength develop-
ment. This is a good agreement with the IR result of GMK2 which
shows the higher value of the wavenumber of the main band
(1094 cm1) and the appearance of a more shoulder band around
1157 cm1. The geopolymer matrix of GMK1 is homogeneous, more
compact and denser matrix due to more depolymerization of
metakaolin, MK1. This may be due to the lower gibbsite content in
Kao1. Berlinite (AlPO4) formed during the geopolymerization with
GMK1 is embedded and dispersed in the geopolymer matrix and
contributes to reinforcing the structure of geopolymer cements.
This is a good agreement with the value of compressive strength of
GMK1 which is higher than the one of the GMK2.
4. Conclusion
Kaolins, Kao1 and Kao2 content about 80.9 wt% and 64.5 wt% of
kaolinite and 11 wt% and 28 wt% of gibbsite, respectively were
transformed to metakaolins, MK1 and MK2, respectively. The X-ray
patterns of MK1 and MK2 show the presence of g-Al2O3 which is
the calcination product of gibbsite. It could be shown that the
calcination product of gibbsite (g-Al2O3) reacts with a phosphoric
acid solution to form berlinite into the structure of phosphate-
based geopolymer cements. The compressive strengths of
phosphate-based geopolymer cements from MK1 and MK2 are
54.41 and 36.01 MPa, respectively. It was typically found that the
286 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288

mechanical and microstructural properties of metakaolin- negatively the compressive strength of metakaolin-phosphate-
phosphate-based geopolymer cement depend on the chemical based geopolymer cements. Regard with berlinite, though it has
and mineralogical compositions of the raw kaolin used. We can similar physical properties than quartz and can play as filler into the
conclude that the presence of gibbsite in the raw kaolin affects matrix, its presence is no longer benefit from strength

Fig. 11. SEM images and EDS results of GMK1 and GMK2.
 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288 287

Fig. 11. (continued).

development. This is because the formation of berlinite involves the References

dealumination of raw aluminosilicate and its condensation with
excess or non-reacted PO3
4 in the matrix due to the low reactivity
[1] J. Davidovits, Geopolymers: inorganic polymeric new materials, J. Therm. Anal.
37 (1991) 1633e1656.
of the aluminosilicate (MK2). [2] F. Zibouche, H. Kerdjoudj, L.J.-B. d'Espinose, H. van Damme, Geopolymers from
Algerian metakaolin. Influence of secondary minerals, Appl. Clay Sci. 43 (2009)
453e458.
[3] C.H. Rüscher, A. Schulz, M.H. Gougazeh, A. Ritzmann, Mechanical strength
Acknowledgement devel- a moment of geopolymer binder and the effect of quartz content, in:
W.M. Kriven, J. Wang, Y. Zhou, A.L. Gyekenyesi, S. Kirihara, S. Widjaja (Eds.),
Herve
Tchakoute

Kouamo gratefully acknowledges The World Developments in Strategic Materials and Computational Design IV, John Wiley
& Sons, Inc., Hoboken, NJ, USA, 2013, http://dx.doi.org/10.1002/
Academy of Sciences (TWAS) for financially support under the 9781118807743.ch2.
grant N 16-039 RG/CHE/AF/AC_I e FR3240293320. [4] Z. Zhang, H. Wang, X. Yao, Y. Zhu, Effects of halloysite in kaolin on the
288 H.K. Tchakout
e et al. / Materials Chemistry and Physics 199 (2017) 280e288
formation and properties of geopolymers, Cem. Concr. Comp. 34 (2012)
709e715.
[5] H.K. Tchakoute, C.H. Ruscher, J.N.Y. Djobo, B.B.D. Kenne, D. Njopwouo, Influ-
ence of gibbsite and quartz in kaolin on the properties of metakaolin-based
geopolymer cements, Appl. Clay Sci. 107 (2015) 188e194.
[6] J. Davidovits, Geopolymer, Chemistry, and Applications, Institute Geopolymer,
Saint-Quentin, France, 2008.
[7] H.K. Tchakoute
, C.H. Rüscher, E. Kamseu, M.C. Bignozzi, C. Leonelli, Influence
of the molar concentration of phosphoric acid solution on the properties of
metakaolin-phosphate-based geopolymer cements, Submitted in Appl. Clay
Sci.
[8] S. Louati, W. Hajjaji, S. Baklouti, B. Samet, Structure and properties of new eco-
material obtained by a phosphoric acid attack of natural Tunisian clay, Appl.
Clay Sci. 101 (2014) 60e67.
[9] W.R. Taylor, Application of infrared spectroscopy to studies of silicates glass
structure: examples from the melilite glasses and the systems Na2O-SiO2 and
Na2O-Al2O3SiO2, Proc. Indian Acad. Sci. Earth planet Sci. 99 (1990) 99e117.
[10] L. Stoch, M. Sroda, Infrared spectroscopy in the investigation of oxide glasses
structure, J. Mol. Struct. 511e512 (1999) 77e84.
[11] D.E.C. Corbridge, E.J. Lowe, The Infra-red spectra of some inorganic phos-
phorus compounds, J. Chem. Soc. (1954) 493e502, http://dx.doi.org/10.1039/
JR9540000493.
[12] J. Davidovits, Geopolymer Chemistry and Applications, third ed., Institute
Geopolymer, Saint-Quentin, France, 2011.
[13] M. O'Hara, J. Duga, H. Sheets, Studies in phosphate bonding, Ceram. Bull. 51
(1972) 590e595.
[14] J. Morris, P. Perkins, A. Rose, W. Smith, The chemistry and binding properties
of aluminum phosphates, Chem. Soc. Rev. 6 (1977) 173e194.
[15] I. Rashkovan, L. Kuzminskaya, V. Kopeikin, Thermal transformations in the
aluminum phosphate binding agent, Neorganicheskie Mater. 2 (1966)
541e549.
[16] A. Chistyakova, V. Sivkina, V. Sadkov, A. Khashkovskaya, L. Povysheva,
Investigation of some aluminum orthophosphates, Neorganicheskie Mater. 5
(1969a) 536e543.
[17] A. Chistyakova, V. Sivkina, V. Sadkov, A. Khashkovskaya, L. Povysheva,
Investigation of an aluminophosphate binder, Neorganicheskie Mater. 5
(1969b) 1573e1580.
[18] J. Ando, T. Shinada, Hiraoka, Reaction of monoaluminum phosphate with
alumina and magnesia, Yoqyo Kyokai Shi 82 (1974) 644e649.
[19] J. Cassidy, Phosphate bonding then and now, Ceram. Bull. 56 (1977) 640e643.
[20] J. Chiou, D. Chung, Improvement of the temperature resistance of aluminum-
matrix composites using and acid phosphate binder, J. Mater. Sci. 28 (1993)
1435e1487.
[21] K. Byrappa, S. Srikantaswamy, G.S. Gopalakrishna, V. Venkatachalapathy, In-
fluence of admixtures on the a-b berlinite transition, J. Mater. Sci. Lett. 5
(1986) 347e348.
[22] W.R. Beck, Crystallographic inversions of the AlP04 polymorphs and their
relation to those of silica, J. Am. Ceram. Soc. 32 (1949) 147e151.
[23] H.K. Tchakoute
, C.H. Rüscher, Mechanical and microstructural properties of
metakaolin-based geopolymer cements from sodium waterglass and phos-
phoric acid solution as hardeners: a comparative study, Appl. Clay Sci. 170
(2017) 81e87.