Applied Clay Science 102 (2014) 213–219

Contents lists available at ScienceDirect

Applied Clay Science

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

Research paper

The identification of geopolymer affinity in specific cases of clay materials

Ivana Perná ⁎, Tomáš Hanzlíček 1, Monika Šupová 1
Institute of Rock Structure and Mechanics, v.v.i., Academy of Sciences of the Czech Republic, 18209 Prague 8, Czech Republic

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

Article history: The article describes a synthesis of a geopolymer matrix based on a specific raw material –thermally treated shale –
Received 21 February 2013 a waste material from mining activity. The intensity of the thermal treatment has been studied by solid-state 27Al
Received in revised form 13 August 2014 nuclear magnetic resonance. The amount of 38–54 wt.% of aluminum ions in tetra-coordination, according to the
Accepted 15 September 2014
sampling point, signals the possibility of converting the waste material into a solid geopolymer matrix. A mineral-
Available online 12 October 2014
ogical analysis has specified that this value corresponds to the aluminum ions involved in two different shale
Keywords:
components: the amorphous clayed residues and a mullite phase. In spite of the high content of mullite and quartz,
Geopolymer the amorphous clay residues in the shale react with an aqueous alkaline solution and create stabile solids which
Raw material have shown excellent mechanical properties. The character of the geopolymer bonds has been investigated by in-
Infrared spectroscopy frared spectroscopy, which has proven the presence of geopolymer chaining and distinguished them from mullite
Shale wavenumbers. The obtained results have confirmed that the shale could be used as a new raw material for
geopolymer synthesis.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction There is a wide, still increasing, spectrum of source materials suitable

Geopolymers are materials known since 1979, when they were first
named by Davidovits (1991). The origin of geopolymers has been
described by many authors (Alonso and Palomo, 2000; Xu and van
Deventer, 2000; Lee and van Deventer, 2003; Davidovits, 2008). Ther-
mally activated alumino-silicate minerals react with a highly alkaline
solution at ambient temperature. One of the conditions of this reaction
is the presence of a tetra- or penta-coordinated aluminum ion ([4] Al3+,
[5] Al3 +), acquired by the thermal activation of clay materials (Sanz
et al., 1988). Besides the thermal activation of clay materials, the effects
of the 27Al coordination changes were identified in the ashes of coal
burning (fluid-burning technology) and in other materials with the po-
sition of [4] Al3+ and/or [5] Al3+ (Phair et al., 2004; Duxson et al., 2007;
Yunsheng et al., 2007; Buchwald et al., 2009). Only a tetra- and penta-
coordinated aluminum ion could be hydrated in an aqueous alkaline so-
lution and could create a connection with the conformable coordinated
silicon through oxygen. This reaction is called geopolymerization and
creates polymeric \Si\O\Al\\ bonds well-ordered in a three-
dimensional network. The ratio of silicon and aluminum tetrahedrons
influences the properties of the geopolymers, e.g. a higher content of sil-
icon tetrahedrons than the classical rate (1:2) in kaolin improves the
compressive strengths of the final solid (Davidovits, 1991, 2008). The
negative charge of an alumina tetrahedron is balanced by the presence
of alkali or alkaline-earth metal ions, commonly Na+, K+ or Ca2+.
⁎ Corresponding author. Tel.: +420 266009253; fax: +420 284680105.
E-mail address: perna@irsm.cas.cz (I. Perná).
1
Tel.: +420 266009253; fax: +420 284680105.
for a geopolymerization including waste materials as well. The raw
materials used the most often are calcinated kaolins (pure and high-
quality) with a significant component of the kaolinite mineral (Duxson
et al., 2007; Davidovits, 2008; Medri et al., 2010; Yunsheng et al.,
2010), coal fly ashes (Phair et al., 2004; Criado et al., 2007) and blast-
furnace slag (Yunsheng et al., 2007; Fu et al., 2011). Chindaprasirt et al.
(2009) prepared coal-bottom-ash geopolymer and Songpiriyakij et al.
(2010) following new trends in the combustion of alternative fuels
made a geopolymer based on biomass ashes. It was also found that
non-used kaolinitic clay in the ceramic industry (Perna et al., 2006)
and ceramic waste material so-called “white water” containing only a
certain proportion of the clay component could be exploited
(Hanzlicek et al., 2006). In many countries the alternative raw materials
for the geopolymerization are still being searched and applied, for exam-
ple natural zeolites (Villa et al., 2010), calcinated paper sludge (Antunes
Boca Santa et al., 2013), rice husk ash (Bouzón et al., 2014), lignite bot-
tom ash (Sathonsaowaphaka et al., 2009), etc.
The essential property of the finally formed 3D net is that this
network could accept many different kinds of additives including
waste materials. The structure and properties of the geopolymer are
affected by the type, quality and amount of the additives. Some (active)
additives, especially ashes and slag, are directly chemically bonded into
the formed net, whereas the others (non-active) are encapsulated into
the geopolymer net (Davidovits, 1991, 2008). Other main properties
of the geopolymer are a high early strength, excellent mechanical
properties, acid and alkali resistance, compact microstructure, low
porosity, fire and heat resistance, low leaching ability and freeze–thaw
resistance (Davidovits, 1991, 2008; Palomo et al., 1999; Roy, 1999; Xu

http://dx.doi.org/10.1016/j.clay.2014.09.042
0169-1317/© 2014 Elsevier B.V. All rights reserved.
214 I. Perná et al. / Applied Clay Science 102 (2014) 213–219
and van Deventer, 2000; Duxson et al., 2005; Yunsheng et al., 2007;
Zhang et al., 2008; Fu et al., 2011).
The properties mentioned above predetermine its practical utilization:
geopolymer cement, construction or composite materials (Perna et al.,
2010), materials for the renovation and restoration of historical monu-
ments (Hanzlicek et al., 2009), fire-resistant materials (Davidovits, 1991,
2008), the fixation of heavy metals (Phair et al., 2004; Yunsheng et al.,
2007), the inhibition of hazardous material (Guangren and Yuxiang,
2002; Zhang et al., 2008; Perna and Hanzlicek, in press) and many other
applications (Perna et al., 2009).
The main aim of this article is to find a new source of a material for a
geopolymerization and to verify properties of a solid geopolymer matrix.
The tested material was the shale from the Zbuch deposit (West Bohemi-
an region, Czech Republic). The coal mining in the area leaves behind a
large number of dumps of the extracted material. This specific material
contained a proportion of coal which has been slowly burning over the
last 50 years. The coal burning was sufficient for the thermal activation
of the shale, which naturally changed the aluminum-ions coordination.
The focus of this work is a verification of the aluminum-ion transforma-
tion and a determination of the possible utilization of this waste material.
2. Materials and experimental procedure
2.1. Materials
This specific waste material formed multiple layers over the coal
veins and was extracted and heaped beside coal-mining installations.
The remnants of coal in this material and the contact of the stacked
material with air, humidity and pressure initiated slow coal burning.
Long-term exposure to elevated temperature naturally changes the
coordination of aluminum ions, like the thermal treatment of clay mate-
rials obtained by clay calcination. The clay substances were hence
“activated” or naturally transformed by the effect of self burning. The
thermal treatment of the clay double-layer minerals generally means
“activation” or shift of the [6] Al3+ ions in their active or meta-stable
[5] Al3 + position to the oxygen. When the changes in aluminum
coordination could not be determined by the controlled temperature,
we labeled the material as naturally activated, even though the activa-
tion was a consequence of a previous coal-mining activity.
The naturally activated shale from the above-mentioned region was
used. In this large deposit area (800 × 600 m), samples were taken in a
quantity of 5 kg from three different parts of the heaped shale. The sam-
pling points were more than 300 m apart, and the points selected also
differed in the color of the material — from light to dark red. The mate-
rial obtained at each sampling point was homogenized and the amount
was reduced by quartering for the subsequent analyses. The average
chemical compositions are presented in Table 1 (L.O.I. — loss on
ignition).
2.2. Sample preparation
The shale was treated in a laboratory dryer for 8 h and then ground
by a Retsch friction mill and later by a vibration mill reaching an average
diameter around 50%: 8.05–9.93 μm.
The 50 g of milled shale was mixed with an alkaline activator
prepared from KOH and soluble potassium silicate. The reaction began
in 15–20 min. The homogenized mixture was pulled into prism molds
(the EN 196-1 standard size of 40 × 40 × 160 mm), which were covered
Table 1
The chemical composition of three shale samples.
Shale/oxide SiO2 Al2O3 CaO MgO Fe2O3 K2O SO3
Sampling point 1 53.75 30.27 0.81 0.68 5.60 2.56 1.44
Sampling point 2 54.47 31.41 0.53 0.57 5.59 2.64 0.73
Sampling point 3 54.69 28.63 0.59 0.62 7.07 3.72 0.73
by a plastic sheet to retard surface-water evaporation. The molds were
placed in the laboratory drier for 3 h (at a temperature of 60 °C). The
times for setting and hardening varied because of the different sampling
points from 24 h to 5 days. After this period, the samples were removed
from the molds and cured at room temperature. The final hardening oc-
curred at laboratory conditions (temperature 25 °C, humidity 55%). The
mechanical properties were determined and XRD and IR analyses con-
ducted after 28 days.
2.3. Methods
The chemical analyses were performed by an X-ray-fluorescence
(XRF) analyzer (Spectro IQ, Kleve, Germany), where the target material
was palladium and the target angle from the central ray was 90°. The
focal point was a 1 mm × 1 mm square, and the maximum Anode
Dissipation was 50 Watts with 10-cfm forced-air cooling. The tested
samples were prepared by the pressed-pellet method.
The solid state MAS–NMR (magic angle spinning–nuclear magnetic
resonance) was measured using a Bruker Avance 500 WB/US NMR
spectrometer (Karlsruhe, Germany, 2003) in 4-mm double-resonance
probeheads at a magic angle spinning (MAS) of ωr/2π = 13 kHz. The
27
Al MAS–NMR spectra were acquired at 130.287 MHz, using a tip
angle of 20° (1 μs pulse length) with a recycle delay of 2 s. The spectra
were referenced to Al (NO3)3 (0.0 ppm).
The X-ray diffraction (XRD) examination of the crystallographic
phases was performed with an X'Pert PRO θ–θ powder difractometer
with parafocusing Bragg-Brentano geometry using Cu Kα radiation
(λ = 1.5418 Å, generator setting: 40 kV, 30 mA). An ultrafast
X'Celerator detector was employed over the angular range 7–70° (2θ)
with a step size of 0.017° (2θ) and a counting time of a 20.00 s step−1.
The data evaluation was performed using the High Score Plus V 2.2e
PANalytical software package, Almelo, Netherlands.
The infrared spectra were measured using the Protégé 460 E.S.P.
(Thermo Nicolet Instruments Co., Madison, USA) spectrometer over
the range between 4000 and 400 cm− 1 at a resolution of 4 cm−1,
averaging 128 scans by the ATR technique to minimize disruption to
the sample caused by the preparation techniques (Rees et al., 2007).
Spectral decomposition (or deconvolution) was performed using the
OMNIC 7 program, with a Gaussian function. The starting calculation
parameters for the curve fitting process, such as the number of
component bands, band positions, width at half-maximum, relative in-
tensities, and shape, were predetermined by the combined procedures
of the secondary derivative method and Fourier self-deconvolution.
The strength tests were confirmed by the European Standard Norm
EN 196-1 (Methods for Testing Concrete, 2005) using prisms with a
standard size of 40 × 40 × 160 mm after 28 days (laboratory
temperature and pressure). The tests were executed on an E156
strength motorized press (strain rate 0.6 ± 0.2 MPa/s) with suitable
devices (producer: Matest company, Italy). At first, the flexural strength
was measured in the three-point bending configuration (support
distance 100 mm) with an E172-01 flexural device. Portions of prism
40 × 40 × 160 mm broken in flexure were used for compressive
strength determination with an E170 compression device. Three prisms
of each sampling point material were tested and the mean and standard
deviation reported. Each value for flexural and compressive strength is
the average of three or six measurements, respectively.
3. Results and discussion
The chemical analyses of the presented samples are very similar. The
low loss on ignition (L.O.I.) in Table 1 shows that the burning of the coal
part in the shale was practically finished. From the chemical composi-
L.O.I.
tion, we could calculate a clayed content of about 49 to 50 wt.%. The
2.98 XRD analysis of the shale (Fig. 1) confirmed a thermal transformation
2.00 of the material and showed quartz Q (SiO2) content as a major phase
1.60
together with mullite M (3Al2O3 · 2SiO2). The minor phase was a
 I. Perná et al. / Applied Clay Science 102 (2014) 213–219
high-temperature formation of quartz (C — cristobalite SiO2). The XRD
pattern also identified traces of hematite H (Fe2O3), mica Mi (KAl2(Al
Si3O10)(OH)2) and dickite D (Al2O3 · 2SiO2 · 2H2O). Traces of clayed
material on the one hand and the formation of mullite on the other
show the differences in the burning temperature at the shale deposit.
The coal substances, randomly distributed, remained in the extracted
material burned partially or completely. Different temperatures create
different crystal phases or low temperature maintains the found clay
mineral. The pattern of XRD also shows the roentgen-amorphous
phase participation.
The highest temperatures of thermal activation could be estimated
from the mullite and cristobalite occurrence. The phase transformations
of mullite and cristobalite are tabulated (Chen et al., 2000; Lee et al.,
2008) at temperatures higher than 1050 °C and depend also on the
iron impurities in the original clay material. Another significant factor
of the transformation occurrence is a dwell at the highest temperature,
which in the case of this type of the shale could not be clearly defined.
The coordination of the aluminum ion of naturally activated shale
hence could not be estimated and must be determined by solid-state
27
Al MAS–NMR. The results are presented in Fig. 2 and Table 2 and
show the dominant types of 27Al-ion coordination.
The results showed that all three materials contain tetra- and hexa-
coordinated aluminum ions whose positions correspond to 58 ppm and
0 ppm, respectively. The content of the tetra-coordinated aluminum ion
is usually significant for the geopolymer synthesis. In the case of
sampling point 1, only 38% of aluminum ions are in tetra coordination.
This amount (38%) of course includes also the aluminum ions presented
in mullite and in clay residues. These residues correspond to the
amorphous part of the tetra-coordinated aluminum ions, which only
takes part in hydration by an aqueous alkaline solution and could create
geopolymer bonds.
In order to verify if the amount of the tetra-coordinated aluminum
ions is sufficient for geopolymer matrix formation the test samples
were prepared (three prisms of each sampling point material). The
strength results of 28-day-old samples are summarized in Table 3.
Each value for flexural and compressive strength is the average of
three or six measurements, respectively. The values in parentheses
correspond to the standard deviation. The table is completed with the
setting and hardening periods of each mixture.
The mixing tests showed that the reaction of naturally activated
shale and an alkaline solution (potassium silicate and potassium
Fig. 1. The XRD analysis of natural shale.
215
hydroxide) creates solid and stable materials. The prepared solid mate-
rials were evaluated from the three points of view:
1. the best combination of flexural and compressive strength,
2. amount of tetra-coordinated aluminum ions, and
3. setting and hardening period.
All prepared solid materials have acceptable compressive strengths
exceeded 48 MPa and in case of sampling point 3 the value reaches
61.48 MPa. The flexural strengths are lower for sampling points 1 and
2 (3.28 MPa and 1.87 MPa, respectively) in comparison with sampling
point 3 (8.90 MPa). What is very important is that these values
correspond to the matrix and could be modified and possibly improved
by selecting the amount and type of additives.
The laboratory tests proved that the sample taken from point 3
contains a sufficient amount of aluminum ions in a position which
allows this material to be used as a raw material for a geopolymer
matrix. This solid material has convenient setting and hardening period
and excellent mechanical properties (Table 3). The materials from
sampling points 1 and 2 could be used only as possible additives. For
the subsequent experiments, the material from sampling point 3 was
selected.
The shale geopolymer matrix (solid matter made from shale) was
analyzed by X-ray diffraction; the resulting pattern is presented in
Fig. 3. The results showed the presence of quartz SiO2 (Q) as a major
phase, mullite 3Al2O3 · 2SiO2 (M), cristobalite SiO2 (C), magnetite
Fe2O3 (Ma) as minor phases and a certain proportion of the roentgen-
amorphous phase. There are also traces of hematite Fe2O3 (H). Figs. 1
and 3 are very similar owing to the dominant content of mullite and
quartz. They should be studied once again and the content of the
aluminum ions participating in the formation of geopolymer netting
determined.
To confirm the geopolymer bonds of the final solid bodies and to
define the character of the original shale and the shale matrix, infrared
spectroscopy was used and the result curves in determinative spectral
regions are presented (Fig. 4).
FTIR spectroscopy of original shale, in the spectral region 3600–
3700 cm− 1 (Fig. 4A), proved a presence of dickite detected by XRD
analysis of shale. Dickite has the same basic layer structure as kaolinite,
but owing to different hydrogen bonding between the layers, dickite
differs from that of kaolinite. Only three resolved bands can be found
in the spectrum of dickite, as a consequence of different geometry of
216 I. Perná et al. / Applied Clay Science 102 (2014) 213–219

Fig. 2. The solid-state 27Al MAS–NMR analyses of shales.

OH\O bonds. The absorption near 3620 cm−1, is assigned to the inner complex bands collected over the 800–1300 cm−1 spectral region
hydroxyl groups. The vibrations of two surface hydroxyls, forming (Figs. 5, 6 and Table 4) and indicates an increase in the substitution of
stronger hydrogen bonds with the adjacent layers, are coupled and the tetrahedral Al in the silicate network due to alkaline activation
give only one band at ~ 3650 cm−1. The third layer-surface hydroxyls and the consequently developing geopolymer network (Lee and van
form weak hydrogen bonds and absorb at ~ 3700 cm− 1 (Madejová, Deventer, 2003). This effect can also be confirmed by the reduction in
2003). the intensity of the spectral shoulder at ~ 560 cm− 1 attributable to
Fig. 4B, in the spectral region 400–1800 cm−1, clearly demonstrates silicates and/or aluminosilicates with long-range structural order in
the presence of vibration bonds, which were codified by many authors the activated shale sample (Sitarz et al., 1997). The alkali-activated sam-
(Shoval et al., 2001; Lee and van Deventer, 2003; Madejová, 2003; ple contains also a small amount of carbonates at ~1400 cm−1 (Fig. 4B).
Criado et al., 2007; Rees et al., 2007; Yunsheng et al., 2007, 2010; The formation of carbonates (the superficial efflorescence) is caused by
Bagchi et al., 2010), confirming the presence of dickite, mullite, the reaction of air humidity and air CO2 content with potassium or sodi-
cristobalite and quartz. As it can be easily seen from Fig. 4B, the position um ions when the ratio of these alkalis to tetra- and penta-coordinated
of the principal Si\O\T asymmetric stretching band located originally Al ions is not defined precisely. Another band confirming the presence
at 1039 cm−1 was found to shift to lower wavenumbers (1007 cm−1).
This shift was proved also by the deconvolution procedure of the
Table 3
The strength results of 28-day-old samples completed with the setting and hardening
Table 2 periods.
The content of tetra- and hexa-coordinated aluminum ions.
Shale Setting and Flexural strength Compressive strength
Shale [4] Al3+ [6] Al3+ hardening period [MPa] [MPa]
Sampling point 1 38.0% 62.0% Sampling point 1 2 days 3.28 (0.24) 48.35 (1.94)
Sampling point 2 38.3% 61.7% Sampling point 2 5 days 1.87 (0.08) 55.05 (1.48)
Sampling point 3 54.1% 45.9% Sampling point 3 24 h 8.90 (0.91) 61.48 (0.88)
 I. Perná et al. / Applied Clay Science 102 (2014) 213–219 217

Fig. 3. The XRD analysis of the shale geopolymer matrix.

of carbonate at 875 cm−1 is probably overlapped by the mullite/dickite
band at 920 cm−1 (Fig. 6).
The XRD analysis of the shale confirmed besides quartz, hematite
and mica that this raw material contains natural (dickite) and high-
temperature forms (cristobalite and mullite) of clay material as a result
of its origin — the slow and unequal burning of coal residues in the
waste material from mining activity. Traces of a dickite in original
shale and their disappearance in the shale geopolymer matrix were co-
incidently demonstrated by XRD and FTIR analyses. Despite the fact that
the 27Al MAS–NMR analyses identified the content of the tetra-
coordinated aluminum ion in all studied samples, the mixing tests ex-
cluded the materials from sampling points 1 and 2 as raw materials
for geopolymer synthesis due to long setting and hardening periods.
The mechanical properties investigation showed that the alkaline acti-
vation of naturally activated shale from sampling point 3 produces a sta-
ble material with favorable flexural and compressive strengths
(8.90 MPa and 61.48 MPa, respectively). In spite of the presence of the
crystalline phases detected by XRD that overlap and influence the
position of the geopolymer peaks to the lower wavenumbers the infra-
red spectroscopy of the shale geopolymer matrix proved an occurrence
of codified vibration bands confirming geopolymer network.
4. Conclusion
In this study, a specific raw material for geopolymer synthesis was
investigated. Shale, the waste material from ancient coal mining activi-
ty, was naturally thermally affected by slow and unequal burning of coal
residues. The results of the XRF analysis confirmed a high proportion of
SiO2, Al2O3 and Fe2O3 and XRD analyses showed a considerable content
of mullite complemented by quartz, cristobalite, hematite, mica, dickite
and roentgen-amorphous phase in all of the tested materials from three
separate sampling points. The 27Al MAS–NMR analyses detected the
total amount of the tetra-coordinated aluminum ions but did not distin-
guish their participation in mullite and in clay residues, only which
could be transformed by an alkaline solution to a geopolymer binder.

Fig. 4. A comparison of the FTIR spectra of the original shale and the shale geopolymer matrix.
218 I. Perná et al. / Applied Clay Science 102 (2014) 213–219

Acknowledgments

This work was carried out thanks to the support of the long-term
conceptual development research organization RVO: 67985891 and
the project of the Ministry of Industry and Trade of the Czech Republic
No. FR-TI2/390.

Fig. 5. The FTIR spectra deconvolution of the original shale.
Fig. 6. The FTIR spectra deconvolution of the shale geopolymer matrix.
The mixing tests proved the reaction of this material with an aque-
ous alkaline solution which creates a solid and stable matrix with differ-
ent setting and hardening period according to the sampling point. The
best geopolymer matrix prepared has flexural strengths of 8.9 MPa,
and the compressive strengths reach the value of 60 MPa.
The dominant crystal phases in shale geopolymer matrix deter-
mined by XRD, especially mullite, quartz, cristobalite and hematite,
could complicate any identification of a geopolymer binder occurrence.
However, FTIR analyses could distinguish between geopolymer and
mullite Al\O bonding and this method proved the identification of
geopolymer bonding in matrix based on shale raw material.
References
Alonso, S., Palomo, A., 2000. Alkaline activation of metakaolin and calcium hydroxide mix-
tures: influence of temperature, activator concentration and solids ratio. Mater. Lett.
47, 55–62.
Antunes Boca Santa, R.A., Bernardin, A.M., Riella, H.G., Kuhnen, N.C., 2013. Geopolymer
synthetized from bottom coal ash and calcined paper sludge. J. Clean. Prod. 57,
302–307.
Bagchi, B., Das, S., Bhattacharya, A., Basu, R., Nandy, P., 2010. Mullite phase enhancement
in Indian kaolins by addition of vanadium pentoxide. Appl. Clay Sci. 47, 409–413.
Bouzón, N., Payá, J., Borrachero, M.V., Soriano, L., Tashima, M.M., Monzó, J., 2014. Refluxed
rice husk ash/NaOH suspension for preparing alkali activated binders. Mater. Lett.
115, 72–74.
Buchwald, A., Hochmann, M., Posern, K., Brendler, E., 2009. The suitability of thermally ac-
tivated illite/smectite clay as raw material for geopolymer binders. Appl. Clay Sci. 46,
303–304.
Chen, C.Y., Lan, G.S., Tuan, W.H., 2000. Microstructural evolution of mullite during the
sintering of kaolin powder compacts. Ceram. Int. 26, 715–720.
Chindaprasirt, P., Jaturapitakkul, C., Chalee, W., Rattanasak, U., 2009. Comparative study
on the characteristics of fly ash and bottom ash geopolymers. Waste Manag. 29,
539–543.
Criado, M., Fernández-Jiménez, A., Palomo, A., 2007. Alkali activation of fly ash: effect of
the SiO2/Na2O ratio: part I: FTIR study. Microporous Mesoporous Mater. 106,
180–191.
Davidovits, J., 1991. Geopolymers: inorganic polymeric new materials. J. Therm. Anal. 37,
1633–1656.
Davidovits, J., 2008. Geopolymer: Chemistry & Application. Geopolymer Institute, Saint-
Quentin, France.
Duxson, P., Provis, J.L., Lukey, G.C., Mallicoat, S.W., Kriven, W.M., van Deventer, J.S.J., 2005.
Understanding the relationship between geopolymer composition, microstructure
and mechanical properties. Colloids Surf. A Physicochem. Eng. Aspects 269, 47–58.
Duxson, P., Fernández-Jiménez, A., Provis, J., Lukey, G., Palomo, A., van Deventer, J., 2007.
Geopolymer technology: the current state of the art. J. Mater. Sci. 42 (9), 2917–2933.
Fu, Y., Cai, L., Yonggen, W., 2011. Freeze–thaw cycle test and damage mechanics models of
alkali-activated slag concrete. Constr. Build. Mater. 25, 3144–3148.
Guangren, Q., Yuxiang, L., 2002. Improvement of metakaolin on radioactive Sr and Cs im-
mobilization of alkali activated slag matrix. J. Hazard. Mater. B92, 289–300.
Hanzlicek, T., Perna, I., Steinerova, M., Straka, P., 2006. Recycling of ceramic wastes
into the geopolymer construction material. In: Önal, G., et al. (Eds.), Proceedings
of XXIII. International Mineral Processing Congress. Promed Advertising Agency,
Istanbul, pp. 2273–2276.
Hanzlicek, T., Steinerova, M., Straka, P., Perna, I., Siegel, P., Svarcova, T., 2009. Reinforce-
ment of the terracotta sculpture by geopolymer composite. Mater. Des. 30,
3229–3234.
Lee, W.K.W., van Deventer, J.S.J., 2003. Use of infrared spectroscopy to study
geopolymerization of heterogeneous amorphous aluminosilicates. Langmuir 19,
8726–8736.
Lee, W.E., Souza, G.P., McConville, C.J., Tarvornpanich, T., Kabal, Y., 2008. Mullite formation
in clays and clay-derived vitreous ceramics. J. Eur. Ceram. Soc. 28, 465–471.
Madejová, J., 2003. FTIR techniques in clay mineral studies. Vib. Spectrosc. 31, 1–10.
Medri, V., Fabbri, S., Dedecek, J., Sobalik, Z., Tvaruzkova, Z., Vaccari, A., 2010. Role of the
morphology and the dehydroxylation of metakaolins on geopolymerization. Appl.
Clay Sci. 50, 538–545.
Oréfice, R.L., Vasconcelos, W.L., 1997. Sol–gel transition and structural evolution on mul-
ticomponent gels derived from the alumina–silica system. J. Sol–Gel Sci. Technol. 9,
239–249.

Table 4
The identification of the bands after the deconvolution procedure.

Wavenumber [cm−1] Assignment Phase References

920 Asymmetric stretching(Si\O\Si and Al\O\Si) Mullite, dickite Madejová (2003); Lee and van Deventer (2003);
Bagchi et al. (2010)
1007 Asymmetric stretching T\O\Si (T_Si or Al) Geopolymer Lee and van Deventer (2003); Criado et al. (2007);
within the TO4 tetrahedra Yunsheng et al. (2010)
1032 Si\O stretching Dickite Madejová (2003)
1070–1080 Asymmetric stretching (Si–O–Si and Al–O–Si) Mullite, dickite Oréfice and Vasconcelos (1997);
Lee and van Deventer (2003); Bagchi et al. (2010)
1095 Si–O stretching Dickite Madejová (2003)
1110–1140 Asymmetric stretching (Si–O–Si) Mullite, dickite Madejová (2003); Lee and van Deventer (2003)
1170–1180 Asymmetric stretching (Al–O–Si) Mullite Bagchi et al. (2010)
1200 Asymmetric stretching (Si–O–Si) Spectral shoulders of the Lee and van Deventer (2003)
principal band at 1070 cm−1
 I. Perná et al. / Applied Clay Science 102 (2014) 213–219
Palomo, A., Blanco-Varel, M.T., Granizo, M.L., Puertas, F., Vazquez, T., Grutzeck, M.W.,
1999. Chemical stability of cementious materials based on metakaolin. Cem. Concr.
Res. 29, 997–1004.
Perna, I., Hanzlicek, T., 2014. The solidification of aluminum production waste in
geopolymer matrix. J. Clean. Prod. http://dx.doi.org/10.1016/j.jclepro.2014.04.043
(Article in press).
Perna, I., Steinerova, M., Hanzlicek, T., Straka, P., 2006. Bonding agent on geopolymer
basis. In: Darve, F., et al. (Eds.), Advances in Geomaterials and Structures AGS'06.
LGC-ENIT, Tunis, pp. 773–777.
Perna, I., Hanzlicek, T., Steinerova, M., Straka, P., 2009. Acoustic absorption of geopolymer/
sand mixture. Ceram.-Silik. 53, 48–51.
Perna, I., Hanzlicek, T., Ertl, Z., 2010. Utilization of biomass ashes for construction pur-
poses. In: Doghri, I., et al. (Eds.), Advances in Geomaterials and Structures AGS'10.
LGC-ENIT, Tunis, pp. 661–666.
Phair, J.W., van Deventer, J.S.J., Smith, J.D., 2004. Effect of Al source and alkali activation on
Pb and Cu immobilization in fly ash based “geopolymers”. Appl. Geochem. 19,
423–434.
Rees, C.A., Provis, J.L., Lukey, G.C., van Deventer, J.S.J., 2007. Attenuated total reflectance
fourier transform infrared analysis of fly ash geopolymer gel aging. Langmuir 23,
8170–8179.
Roy, D.M., 1999. Alkali-activated cements — opportunities and challenges. Cem. Concr.
Res. 29, 249–254.
Sanz, J., Madani, A., Serratosa, J.M., Moya, J.S., Aza, S., 1988. Aluminium-27 and silicon-29
magic-angle spinning nuclear magnetic resonance study of the kaolinite–mullite
transformation. J. Am. Ceram. Soc. 71, 418–421.
219
Sathonsaowaphaka, A., Chindaprasirt, P., Pimraksab, K., 2009. Workability and strength of
lignite bottom ash geopolymer mortar. J. Hazard. Mater. 168, 44–50.
Shoval, S., Boudeulle, M., Yariv, S., Lapides, I., Panczer, G., 2001. Micro-Raman and FT-IR
spectroscopy study of the thermal transformation of St. Claire dickite. Opt. Mater.
16, 319–327.
Sitarz, M., Mozgawa, W., Handke, M., 1997. Vibrational spectra of complex ring silicate an-
ions — method of recognition. J. Mol. Struct. 404, 193–197.
Songpiriyakij, S., Kubprasit, T., Jaturapitakkul, C., Chindaprasirt, P., 2010. Compressive
strength and degree of reaction of biomass- and fly ash-based geopolymer. Constr.
Build. Mater. 24, 236–240.
Villa, C., Pecina, E.T., Torres, R., Gómez, L., 2010. Geopolymer synthesis using alkaline ac-
tivation of natural zeolite. Constr. Build. Mater. 24, 2084–2090.
Xu, H., van Deventer, J.S.J., 2000. The geopolymerisation of alumino-silicate minerals. Int.
J. Miner. Process. 59, 247–266.
Yunsheng, Z., Wei, S., Qianli, Ch., Lin, Ch., 2007. Synthesis and heavy metal immobilization
behaviors of slag based geopolymer. J. Hazard. Mater. 143, 206–213.
Yunsheng, Z., Wei, S., Zongjin, L., 2010. Composition designe and microstructural charac-
terization of calcined kaolin-based geopolymer cement. Appl. Clay Sci. 47, 271–275.
Zhang, J., Provis, J.L., Feng, D., van Deventer, J.S.J., 2008. Geopolymers for immobilization of
Cr6+, Cd2+, and Pb2+. J. Hazard. Mater. 157, 587–598.