Construction and Building Materials 25 (2011) 22062213

Contents lists available at ScienceDirect

Construction and Building Materials

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

Properties of metakaolin geopolymer hardened paste prepared

by high-pressure compaction
Vladimr Zivica a,, Svetozar Balkovic b, Milan Drabik b
a
b

Institute of Construction and Architecture of the Slovak Academy of Science, Dbravska cesta 9, 84503 Bratislava, Slovak Republic
Institute of Inorganic Chemistry of the Slovak Academy of Science, Dbravska cesta 9, 84536 Bratislava, Slovak Republic

a r t i c l e

i n f o

Article history:
Received 26 November 2009
Received in revised form 12 November 2010
Accepted 14 November 2010
Available online 18 December 2010
Keywords:
Metakaolin
Geopolymer
Compressive strength
Pore structure parameters
Thermal analysis
EDX results

a b s t r a c t
The paper presents the results of the study on the effects of the combination of the low liquid/solid (l/s)
ratio and pressure compaction of the fresh pastes on the properties of the hardened metakaolin geopolymer paste. It is well known that the combination gives the possibility to prepare cement composites
resulting in the excellent engineering properties. The results obtained shown the high dense nano- or
near-nano-pore structure with high degree of its homogeneity and high strength of metakaolin geopolymer hardened paste prepared under the use of very low l/s ratio and pressure compaction.
Two factors were found responsible for the high positive effects:
(i) Mechanical effect of the densifying of the starting mixture and the decrease of the initial porosity
due to action of the applied pressure compaction.
(ii) The well known very positive effect of the l/s ratio decrease on the rate and level of the development of the properties of cement composites.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
It is known that alkali activated aluminosilicates are able to
produce alumino-silicate polymers geopolymers. The hardening
mechanism involves the chemical reaction of geopolymeric
precursors, such as alumino-silicate oxides, with alkali polysilicates
yielding polymeric SiOAl bonds. The production of geopolymeric
precursors is carried out by calcinations of aluminosilicates natural
clay materials. Their source can be also some industrial aluminosilicate waste materials. The result of the hardening mechanism is a
three dimensional zeolitic framework unlike traditional hydraulic
binders in which hardening is the result of the hydration of calcium
aluminates and silicates [17]. This circumstance is a cause of
signicant differences in the quality and variety of the engineering
properties of the composites based on geopolymer and current
cements.
As it is very well known the strength and the level of the other
engineering properties of the cement based materials are signicantly dependent on the w/c ratio used. Also the fact that their
strengths, durability and the quality of other engineering properties are increased when the w/c values are decreased [810]. A
signicant factor limiting the effective use of these positive effects
is the decreasing of the quality of the properties of the composites
due to the decrease of the workability of the fresh composite
mixtures prepared with very low w/c ratios.
Corresponding author. Tel.: +421 2 59309257.
E-mail address: usarziv@savba.sk (V. Zivica).
0950-0618/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2010.11.004

A very attractive solution of this approach represents the

combination of super low w/c ratios, under w/c 0.1 and lower,
and the pressure compaction of the fresh composite mixtures.
The result is a submicroscopic pore structure and an adequate high
increase of the quality of engineering properties of the composites
[1117]. It appears that low-porosity cement composites have a
great potential of reconsideration and modication of composition
and structure.
The aim of the presented work was to study the possibilities of
the use of effects of the application of very low w/c ratio and pressure compaction in the geopolymer cement systems. As a model
the alkali activated metakaolin system was used.

2. Experimental
2.1. Materials
The metakaolin used was the product of kaolin heat-treatment at 650 C for 4 h.
The data on the chemical composition and specic weight of the original kaolin and
metakaolin are given in Table 1. With the aim to compare the reactivity of both
materials also their solution heat was estimated using the method of the kinetics
of its development [18,19].
The comparison of the chemical composition shown an increase in the SiO2,
Al2O3 and Fe2O3 contents of the metakaolin as a direct consequence of the decrease
of the ignition loss due to the heat-treatment of kaolin. Further interesting effects
could be observed: a signicant (ca. 20 times) increased solution heat and, on the
contrary, a decreased specic weight of the metakaolin. The increased solution heat
indicated the increased reactivity of metakaolin opposite to the kaolin. The found
specic weight decrease indicated the decreased compactness of metakaolin. This

V. Zivica et al. / Construction and Building Materials 25 (2011) 22062213

Table 1
Chemical composition and properties of kaolin and metakaolin used.
Chemical composition (%)
Components

Kaolin

Metakaolin

Humidity
Ignition loss
SiO2
A12O3
Fe2O3
CaO
MgO
Specic weight (kg m
Solution heat (J g 1)

0.94
12.20
48.06
36.76
1.42
0.33
0.24
2631
278.2

0.35
1.95
53.68
42.00
1.70
0.21
0.10
2582
5563

2207

effect was conrmed by the porosimetry results showing 31.8% increase of pore
volume and 20.2% increase of total porosity. Further, the found 40.7% decrease of
macro-pore content and 689.8% increase of specic surface area indicating the
metakaolin as a material signicantly ner than the starting kaolin (Table 2).
All the mentioned effects were undoubtedly the consequence of the process of
the heat-treatment producing the amorphous or near-amorphous SiO2Al2O3
Fe2O3 product. Its presence in the metakaolin is conrmed from the DTA-curves
by a very broad endotherm between ca. 200 and 950 C, instead of the sharp endotherm peak at 511 C visible on the DTA-curve of the kaolin (Fig. 1).
According to the results the used metakaolin was the material with SiO2/Al2O3
molar ratio 2.22 showing signicantly increased neness and solution heat both
contributing to the increase of its reactivity as a partner in the process of the geopolymer formation [7,20,21].

2.2. Test specimens and their testing

Table 2
Pore structure parameters of kaolin and metakaolin used.
Pore structure parameter

Kaolin

Metakaolin

Pore volume (mm3 g 1)

Macro-pore portion (%) r > 7500 nm
Pore median (nm)
Total porosity (%)
Specic surface area calculated from
porosimetry data (m2 g 1)

771
26.8
1102
77.8
4.9

1016 (+31.8%)
15.5 ( 40.7%)
772 ( 29.9%)
93.5 (+20.2%)
38.7 (+689.8%)

For the study the hardened alkali activated metakaolin pastes prepared with
activator solution/metakaolin ratio (l/s) 0.08 and compacted by the pressure
300 MPa were used. The prepared test specimens were 20 mm-edge cubes. The reference test specimens were prepared with l/s 0.70 using the compaction by hand.
The alkali activator sodium hydroxide was added to the mixture in the form of solution in the adequate quantity of water. The portion of the activator corresponded to
7 wt.% of the weight of the metakaolin.
The preparation of the fresh mixtures represented the addition of the given
activator solution amount to the metakaolin and intensive mixing for 3 min using
electrical mixer. The pressure compaction followed after this procedure. The

Fig. 1. Thermograms of kaolin and metakaolin used.

V. Zivica et al. / Construction and Building Materials 25 (2011) 22062213

2208

20  20  20 mm cube forms were lled by the metakaolin mixture and exposed to

the pressure of 300 MPa with the endurance of 1 min. The preparation of the
reference test specimens was the same but under the compaction by hand.
The prepared test specimens hardened 24 h in the forms, at temperature 20 C
and relative humidity 95%. Then, after of the demoulding, their testing followed.
The test specimens were dried at 105 C and these properties were observed: bulk
weight and compressive strength, hydration products occurred (DTA, TG, EDX), pore
structure (intrusion mercury porosimetry) and morphology of the matrix (scan
microscopy), water bound and the total ignition loss (TG).
2.3. Methods
For bulk weight and compressive strength tests the current testing methods
were used.

Thermal analysis (simultaneous TG and DTA-curves) was carried out on the

equipment SDT 2960 device T.A. Instruments: the sample mass 20 mg, heating rate
was 10 C min 1, ramping from ambient temperature to 1000 C in an air atmosphere. TG mass losses and DTA effects were analysed using the T.A.I. Thermal Analyst Package.
Pore structure was analysed using the intrusion mercury porosimeter mod. 2000
Erba Science working to the pressure 200 MPa with macro-pore unit 120, under the
use of the contact angle 141.3 and surface tension of mercury 0.48 N m 1.
The morphology of the samples was studied by scan electron microscopy using
a Carl Zeiss EVO 40 HV microscope. Before the scanning process, all samples were
coated with gold to enhance the electron conductivity. The samples were also
examined by energy dispersive X-ray analysis (EDX) with spectrometer Quantax
400 to analyse chemical composition.

Table 3
Bulk weight and compressive strength of metakaolin geopolymer pastes and pore structure parameters after 24 h of the hardening.

Pressure
compaction
Reference

Bulk weight
(kg m 3)

Compressive strength
(MPa)

Total porosity
(%)

1862

146.6

15.7

0.03

54.5

814

Pore median
(nm)

Specic surface area

(m2 g 1)

Portion of pores

59

5.5

96

814

5.8

93

Fig. 2. Histogram and pore size distribution curve of metakaolin geopolymer pressured composite.

r < 100 nm
(%)

r > 100 nm
(%)

V. Zivica et al. / Construction and Building Materials 25 (2011) 22062213

2209

Fig. 3. Histogram and pore size distribution curve metakaolin geopolymer reference composite.

3. Results and discussion

3.1. Mechanical properties and pore structure results

Fig. 4. SEM image of the pressure compacted metakaolin geopolymer composite

after 24 h of the hardening.

The results of the estimation of bulk weight, compressive

strength and pore structure parameters are summarized in Table 3.
The signicant difference between the pressure compacted and
reference pastes is evident. An increase in the benet of pressure
compacted paste is 128.7% at the bulk weight and ca. 500 times
at the compressive strength values.
An adequate relationship values of total porosity and pore
median of the pressure compacted probes opposite to those of
reference paste occured (Table 3). At the same time corresponding
content of macro-pores (r > 100 nm) is 4% at the pressure compacted paste and 93% at the reference one, and a reverse ratio
96% and 4%, at micropores contents (r < 100 nm).
The histograms and pore size distribution curves shown in Figs.
2 and 3 provide a suitable illustration about the character of pore
structure. At the comparison it can be seen at the pressure compacted paste a signicant shift of pore size curve and the character

2210

V. Zivica et al. / Construction and Building Materials 25 (2011) 22062213

Fig. 5. Composition of the pressure compacted metakaolin composite after 24 h of the hardening analysed by EDX.

the Grifths law it can cause an increase of the critical tension

in the benet of the pressure compacted paste [22].
The found signicantly high compressive strength of pressure
compacted paste is undoubtedly based on its advanced pore structure developed under the use of the low l/s ratio and pressure compaction. The metakaolin paste prepared under these conditions
was able to overcome that prepared under the current conditions.
It means under the use of the current l/s ratio and without the
pressure compaction.
As it is reported in the literature the currently prepared metakaolin pastes reached the compressive strength 5 MPa after 1 day (l/s
ratio 0.70) and ca. from 15 to 90 MPa after 7 days of the hardening
[23,24]. The compressive strength values between 0.40 to 64 MPa
and 38 MPa are also reported [2527].
3.2. SEM and EDX results

Fig. 6. SEM image of the reference metakaolin geopolymer composites after 24 h of

the hardening.

of the histogram in the area of nano-pores with radius under

100 nm. It shows the pressure compacted paste as a material with
the increased pore structure homogeneity and much ner pore
structure. It may have an interesting and signicant consequence
for the strength characteristics of the material. In the sense with

The shown difference in the comparison of these values with

that 146.6 MPa reached by the pressure compacted paste the
positive character of pore structure was conrmed by the SEM
images (Figs. 4 and 6). As it can be seen the microstructure for
the reference paste is highly in compact, with the large grains
and pores shown as dark places. The occurrence of ake-like layer
structure of metakaolinite particulates is an evidence of the low
development of the geopolymerization process. Evidently, only a
small quantity of the geopolymer product is taking the place at
the surface layer of the metakaolin particulates [28,29].

Fig. 7. Composition of the reference metakaolin geopolymer composite after 24 h of the hardening analysis by EDX.

V. Zivica et al. / Construction and Building Materials 25 (2011) 22062213

Table 4
Composition of the metakaolin geopolymer pastes after 24 h of the hardening
analysed by EDX.
Element as oxide
(wt.%)

Pressure compacted
paste

Reference
paste

Metakaolin

SiO2
A12O3
Na2O
SiO2/A12O3 molar
ratio
Na2O/A12O3 molar
ratio

37.32
33.41
8.38
1.90

49.71
39.44
6.60
2.14

53.68
42.00

2.17

0.41

0.26

Due to the use of the different magnication the comparison of

the SEM results enabled only a limited information on the inuence of the combination and the low l/s ratio and high-pressure
compaction on the morphology of the particles of the system.
But in principle SEM results shown the same tendency of the inuence like it can be seen at the porosimetry results.
The microstructure of the pressure compacted paste is distinctly different. In this material, a grain structure is still apparent,
but the intervening material is of more homogeneous nature, with
small pores. This observation was linked to a strong correlation in
the compressive strength increase found but also reported in the
paper [22].
The coupling of the SEM with an EDX provided an analysis of
the surface of the studied materials. It should be expressed that
the data obtained have only semiquantitative value (Figs. 5 and
7). The presence of the main elements O, Si, Al, Na and minor elements K, Fe and Mg can be seen in the EDX spectra of the geopolymers. Table 4 shows the composition of the geopolymers occurring
in the studied pastes calculated using EDX data. The decreased SiO2
and Al2O3 contents in both pastes opposite to the starting values

2211

are expressed at the pressure compacted paste. The principle of

the effect is not clear. It can be only assumed as a consequence
of the effect of the geopolymer formation.
The content of Na2O in the pressure compacted paste is by
30 wt.% higher opposite to that of reference paste. One reason could
be the different concentrations of the used activator solutions given
by the different l/s ratios. These caused that under the use of the
same NaOH quantity of 7 wt.% but at the different l/s ratios the concentration of the activator solution was 46.7 wt.% at the pressure
compacted paste but only 9.1 wt.% at the reference one.
The signicance of the NaOH concentration in the process of
Connor regeopolymer formation is unquestioned. Rowles and O
ported that the compressive strength of geopolymer depends signicantly also on the Na:Al ratio together with Si:Al. The
compressive strength is increased as the both ratios are increased
[25]. It is well known that variation of the SiO2/Na2O ratio signicantly modies the degree of polymerization of the dissolved species in an silicate solution [24]. This plays a signicant role in
determining the structure and properties of geopolymer gels. In
the case of the NaOH activator with low concentration, there would
be insufcient amount of OH ions to completely dissolve Si4+ and
Al3+ from metakaolin, and insufcient Na1+ to allow for complete
polymerization of the network. These both lead to unreacted
metakaolin part and, therefore, lower strength of material. In the
case the high sodium content, there would be excess OH 1 ions
allowing for a complete dissolution of Si4+ and Al3+ from
metakaolin.
The signicance of the concentration of NaOH solution for the
mechanical properties of the geopolymers is conrmed also by
the results reported in the paper [29]. An increase of a compressive
strength from ca. 35 to 65 MPa has been shown when the NaOH
solution concentration increased from 4 to 12 mol L 1. This effect
is attributed to the enhanced dissolution of the metakaolin

Fig. 8. Thermogravimetric (TG) and differential thermal (DTA) analysis curves of the pressure compacted metakaolin geopolymer composite after 24 h of the hardening.

V. Zivica et al. / Construction and Building Materials 25 (2011) 22062213

2212

Fig. 9. Thermogravimetric (TG) and differential thermal (DTA) analysis curves of the reference metakaolin geopolymer composite after 24 h of the hardening.

particulates and hence the speeded condensation of the monomer

in the presence of NaOH with higher concentration.
3.3. Thermal analysis results
The results of the thermal analysis of studied materials are
shown in Figs. 8 and 9. As it can be seen these results enable only
limited information on the phase composition of system. It is evidently the consequence of the amorphous or near-amorphous character of the constituents of the materials studied, thus a use of any
method is rather restricted. DTA-curves are characterized (i) by the
occurrence of the effects in the region about 100 C which are due to
the loss of evaporable water, and (ii) by the occurrence of the exotherms at 970 C which are caused by the mullite formation. TG
curves shown in both pressure compacted and reference pastes
very near levels of weight loss 7.2 and 7.9 wt.%. Besides the adsorbed water, the loss is also due to the chemically bound water
typical for a series of hydraulic hydrates. The thermoanalytical data
strongly favor a qualitative similarity of phase composition of pressure compacted and reference probes, thus supporting the data
interrelations of pore structure and mechanical tests on the densication and strength development as a unique effect of pressure
compaction.
4. Conclusion
The combination of the low alkali activator solution:binder ratio (l/s) with the pressure compaction of the fresh mixture as a signicant accelerator of the development of the pore structure and
on it dependent strength has been shown. At the given case the
use of l/s 0.08 and 300 MPa compaction pressure resulted in the
very dense near-nano-pore structure with the high degree of the
homogeneity and the compressive strength overcoming the reference material ca. 500 times.

The found positive effect of the combination of the low l/s ratio
and high-pressure compaction also at the another alkali activated
alumino-silicate materials can be expected.
The Si:Al ratio in the starting alumino-silicate materials, neness, the portion and type of alkali activators used, curing regime,
and other parameters are undoubtedly important for the development of the properties of the geopolymer composites with the potential in the quality differences of the systems based on the
various aluminosilicates. Therefore, these factors are worthy of
the more detailed study because the pressured geopolymer composites seem to be very interesting materials.
Acknowledgement
The authors would like to thank to Slovak Grant Agency VEGA
for its support towards this work (Grant No. 2/0055/08).
References
[1] Davidovits J. Geopolymer cements to minimize carbon-dioxide greenhousewarming. Ceram Trans 1993;37:16582.
[2] Davidovits J. Geopolymers: inorganic polymeric new materials. J Therm Anal
1991;37:163356.
[3] Wang Shao-Dong, Xin-Cheng Pu, Scrivener KL, Pratt PL. Alkali-activated slag
cement, concrete: a review of properties and problems. Adv Cem Res
1995;7(27):93102.
[4] Della Roy M. Alkali-activated cement, opportunities and challenges. Cem Concr
Res 1999;29(2):24954.
[5] Duxson P, Fernndez-Jimnez A, Provis JJ, Lukey GC, Palomo A, van Deventer
JSJ. Geopolymer technology: the current state of the art. J. Mater Sci. 2007;42:
291733.
[6] Pacheco-Torgal F, Castro-Gomes J, Jalali S. Alkali-activated binders: a review.
Part 1. Historical background, terminology, reaction mechanism and hydration
products. Constr Build Mater 2008;22(7):130514.
[7] Pacheco-Torgal F, Castro-Gomes J, Jalali S. Alkali-activated binders: a review.
Part 2. About materials and binders manufacture. Constr Build Mater
2008;22(7):131522.
[8] Schulze J. Inuence of watercement ratio and cement content of the
properties of polymer modied mortars. Cem Concr Res 1999;29:90915.

V. Zivica et al. / Construction and Building Materials 25 (2011) 22062213

s law. Cem Concr Res
[9] Naragaj TS, Zahida Banu TS. Generalization of Abram
1996;26(6):93342.
[10] Appa Rao G. Role of waterbinder ratio on the strength development in
mortars incorporated with silica fume. Cem Concr Res 2001;31:4437.
[11] Freidin C. Cementless pressed blocks from waste products of coal-ring power
station. Cem Concr Res 2007;21:128.
[12] Xu Zhongzi, Deng Yongmei, WU Xuequan, Tang Mingshu. Inuence of various
hydraulic binders of performance of very low porosity cementitious systems.
Cem Concr Res 1993;23:46270.
[13] Xu Zhongzi, Tang Mingshu, Beaudoin JJ. Relation between composition
structure and mechanical properties of very low porosity cementitious
systems. Cem Concr Res 1993;23:18795.
[14] Roy DM, Gouda GR, Borovsky A. Very high strength cement pastes prepared by
hot pressing and other high pressure techniques. Cem Concr Res 1972;2:
80720.
[15] Della Roy M, Gouda GR. High strength generation of cement pastes. Cem Concr
Res 1973;3:349.
[16] Xu Zhongzi, Tang Mingshu, Beaudoin JJ. An ideal structure model for very low
porosity cementitious systems. Cem Concr Res 1993;23:37786.
[17] Lin Feng, Meyer Ch. Hydration kinetics modeling of Portland cement
considering the effects of curing temperature and applied pressure. Cem
Concr Res 2009;39(4):25565.
[18] Jambor J. The principle of the pozzolan activity and a new method of its
estimation. Stavebnicky Casopis 1996;9(9):54157, (10):62849 [in Slovak].
[19] Jambor J. Une nouvelle mthode de determination de lactivit pouzzalanique.
Revue Des Matriaux 1962;(564):24056.

2213

[20] Granizo ML. Activation alcalina de metakaolin: desarrollo de nuevos materials.

PhD Thesis. University Autonoma of Madrid; 1998.
[21] Weng L, Sagoe-Crentsil K, Brown T, Song S. Effects of aluminates on the
formation of geopolymers. Mater Sci Eng 2005;117:1638.
[22] Grifth AA. The phenomena of rupture & ow in solids. Philos Trans R Soc
London Ser A Math Phys Sci 1920;A221:16397.
[23] Buchwald A, Hilbig H, Ch Kaps. Alkali-activated metakaolinslag blends
performance and structure in dependence of their composition. J Mater Sci
2007;42:302432.
[24] Duxson P, Mallicoat SW, Lukey GC, Kriven WM, Deventer JSJ. The effect of
alkali and Si/Al ratio on the development of mechanical properties of
metakaolin-based geopolymer. Colloids Surf A Physchem Aspects
2007;292(1):820.
[25] Rowles M, OConnor B. Chemical optimization of the compressive strength of
aluminosilicate geopolymers synthesised by sodium silicate activation of
metakaolinite. J Mater Chem 2003;13:11615.
[26] Kong DLY, Sanjayan JG, Sagope-Crenstil K. Comparative performance of
geopolymers made with metakaolin and y ash after exposure to elevated
temperature. Cem Concr Res 2007;37(12):15836.
[27] Zivica V. Effects of the very low water/cement ratio. Constr Build Mater
2009;23:357982.
[28] Wang H, Li H, Yan F. Synthesis and mechanical properties of metakaolinitebased geopolymer. Colloids Surf A Physchem Eng Aspects 2005;268:16.
[29] Valeria FFB, Kenneth JDM, Clelio T. Synthesis and characterization of materials
based on inorganic polymers of alumina and silica: sodium polysialate
polymers. Int Inorg Mat 2000;2:30917.