Nanoindentation Study of Na-Geopolymers
Exposed to High Temperatures
I. Beleña and W. Zhu
*
Abstract. This paper reports the usefulness of nanoindentation as a characterization and monitoring tool for studying thermal behaviour of Geopolymer materials.
The influence of the manufacturing process of Na-Geopolymers in their micromechanical properties and thermal behaviour has been studied. Two types of
metakaolin-based geopolymer panels with almost identical composition were prepared by injection and pouring methods. Micro-mechanical properties of the two
samples exposed to high temperatures up to 1000 ºC were studied by nanoindentation technique, supplemented by X-ray diffraction (XRD), Nuclear magnetic resonance (NMR), Thermogravimetric analysis (TGA) and Microscopy. Remarkable
differences in micro-mechanical properties and thermal behaviour between the
two samples were found. Statistical nanoindentation has been successfully used to
provide information about the micro-mechanical properties of different phases in
the material and their volume distributions.
1 Introduction
Geopolymers are ceramic-like materials obtained by alkali activation of aluminosilicates raw materials at high pH environment, atmospheric pressure and temperatures below 100 ºC. These materials are inorganic polymers consisting in 3D
amorphous networks of SiO4 and AlO4 tetrahedron alternately linked by sharing all
3+
the oxygen atoms. The Al forth-coordinated is charge balance achieved by the
+
+
presence of alkali cations (usually Na or K ). Their properties are much influenced by the Si/Al ratio in the final network. The starting aluminosilicate materials are disaggregated in a high alkali media and the SiO4 and AlO4 oligomers
(dimers, trimers…) condense giving as a result an amorphous aluminosilicate
polymer and water [1].
I. Beleña
Technological Institute of Construction of Valencia (AIDICO), Technological Park,
Valencia, Spain
e-mail: irene.belenya@aidico.es
W. Zhu
Advanced Concrete and Masonry Centre, University of the West of Scotland, Scotland, UK
e-mail: Wenzhong.Zhu@uws.ac.uk
170
I. Beleña and W. Zhu
Due to the low energy requirements in the production and good mechanical
performance, thermal behaviour and durability, Geopolymer has attracted increasing interest as an ecologically friendly fireproof building material. Many studies
about the thermal behaviour of geopolymers with sodium or potassium silicate
have been reported [1-4]. The metakaolin-based geopolymers (with a potassiumcontaining activator) have shown to achieve thermal stability up to 1200-1400ºC.
The chemical composition, microstructure and macro-mechanical properties have
been widely studied. No attempt, however, has been made to determine the micromechanical properties and their changes in different conditions for these materials.
The aim of the present work was to study the micro-mechanical properties of
Na-geopolymers using nanoindentation technique, which has been used to study
properties of nano/micro-scale features (e.g. Elastic modulus and Hardness) in
many different materials, including composites or multiphase materials [5-8]. In
this study, geopolymers with the same starting materials and composition were
made using two different processes, i.e. direct pouring and injection. Micromechanical properties of the two samples and their changes due to high temperature exposure up to 1000 ºC were studied by Nanoindentation and supplemented
by Microscopy, XRD, MAS-NMR and TGA.
2 Experimental
2.1 Materials and Specimens
Geopolymers tested were synthesised using Metakaolin (Sigma & Aldrich Chemistry, S.A.), Sodium Silica (Massó y Carol, S.A.) and Sodium Hydroxide (Prolabo,
S.A.) (Table 1). The specimens were prepared using a previous optimized composition [9] and different manufacturing processes (Table 2).
Table 1 Raw material composition and physical properties
% oxides (wth) SiO2
Al2O3 Fe2O3 Na2O K2O
3
P.C* d (g/cm )
2
d50 (µm)
m) BET (m /g)
Metakaolin
52.4
43.8
1.4
0.0
0.5
0.0
2.78
5.78
9.0
Na2SiO4.xH2O
27
0.0
0.0
10.6
0.0
62.4
1.6
---
---
Table 2 Details of specimens used
Manufacture process
Sample Moulding
Curing
Chemical composition
Drying
SiO2/Al2O3 Na2O/Al2O3 H2O/ Na2O %Solid
o
~25 C
NaMk-I Injection Closed mould
Two weeks
o
~25 C
o
~25 C
In the air
3.5
0.8
13
34
3.5
0.8
15
36
o
~25 C
NaMk-P Pouring Open mould In absorbent
material
24h
Nanoindentation Study of Na-Geopolymers Exposed to High Temperatures
171
Fig. 1 Optical images of the typical tested specimens (intent spacing being 30 µm): a)
NaMk-I unheated, b) NaMk-P unheated
2.2 Testing Details
The methodology and operating principle for the nanoindentation technique have
been reviewed and presented in detail elsewhere [5-8]. Briefly, the test consists of
making contact between a sample surface and a diamond indenter of known geometry, followed by a loading-unloading cycle while continuously recording the
load, P, and indentation depth, h. The P-h curve obtained is a finger print of the
mechanical properties of the test area. Most commonly, the elastic modulus (E)
and hardness (H) of the test area are determined by analysing the initial part of the
unloading data according to a model for the elastic contact problem. For studying
multiphase composite materials, a refined statistical indentation method has been
used, which involves testing and statistically analyzing a large number of indentation points within a representative sample area [7-8]
The nanoindentation apparatus used in this study was Nanoindenter XP with a
Berkovich indenter. In this study, all testing was programmed in such a way that
the loading started when the indenter came into contact with the test surface and
the load maintained for 30 seconds at the pre-specified maximum value before
unloading. In order to provide statistical analysis of the micro-mechanical properties of different phases in the specimen, a grid of 8x15 indentation points with a
indent spacing of 30 µm was selected, as shown in Fig.1. For studying the effect of
high temperature exposure on properties of the geopolymers, selected samples
were placed in an electrical furnace at various temperatures up to 1000ºC. The
temperature was set to increase at 1ºC/min until 100ºC and 2ºC/min until the
specified temperature and then maintained for 24 hours before natural cooling.
3 Results and Discussions
3.1 Micromechanical Properties of the Geopolymers before
Heating
The results obtained from the statistical indentation test of the two geopolymer
samples are presented in Table 3. Generally, for both samples, three peaks were
172
I. Beleña and W. Zhu
observed in the frequency-mechanical property (i.e. E and H) plots, which led to
the determination of E and H values and volume fractions for the three different
phases in each sample. Microscopic (optical and electron) analysis of the indented
area has been used to identify the nature of these phases. In Fig.1 the grey background mass is the geopolymer matrix, the brightest irregular shaped pieces are
the quartz (present in metakaolin as an impurity) and the slightly bright/white
pieces are those incompletely reacted metakaolinite raw material.
Table 3 Micromechanical properties and volume fractions of phases present in test
samples
Phase
NaMk-I
NaMk-P
E, GPa
H, GPa
V%
E, GPa
H, GPa
Geopolymer (GP)
14
0.5
89
7
0.2
V%
83
Metakaolinite (Mk)
25
1.0
7
12
0.5
13
Quartz (Q)
99
15
4
96
14
4
Results in Table 3 appear to indicate that the micromechanical properties of the
geopolymeric matrix are significantly higher in NaMk-I than in NaMk-P sample,
as well as there is a higher amount of unreacted metakaolinite in NaMk-P than in
NaMk-I. This is probably due to the difference in curing and drying of the two
processes. For NaMk-P the curing time was short (24 h) and in open mould with
water likely evaporating from the top surface, while the NaMk-I was cured in
closed mould for two weeks, thus likely leading to more complete polymerization.
3.2 Thermal Behaviour Study of Geopolymer Samples
Nanoindentation tests were carried out on selected geopolymer samples which had
undergone exposure at various high temperatures (Table 4). They show a moderate increase in the mechanical properties of the geopolymer matrix and partially
reacted metakaolin phases in both specimens with the rising exposure temperature
up to 400ºC. It is believed that such changes are probably due to the dehydration
of both phases, together with a progressive increase in the development of new
bonds (as suggested by the thermal analysis results). The quartz phase remained
o
o
unchanged until 400 C but disappeared at 700 C. There seemed to be little change
in the GP and the MK phases between 400ºC and 700ºC. Dramatic changes in the
sample were observed at high temperature exposure between 700 to 1000ºC. The
mechanical properties of the GP phase in both specimens were more than tripled,
likely due to a partial softening of geopolymer structure from 788ºC and forming a
o
vitreous phase upon cooling, as suggested by the TDA results. In the 1000 C sample, the MK phase was no longer present, but a new phase, Mullite was observed
in both samples. Nepheline was also found in the NaMk-I..
o
Optical images of the 1000 C specimens (Fig.2) revealed that no MK and quartz
phases were present in both samples, and many pores (up to 150 um) were developed in NaMk-I sample. Different phase identification was support by XRD. It is
Nanoindentation Study of Na-Geopolymers Exposed to High Temperatures
173
Table 4 Properties of different phases in geopolymer samples heated to high temperatures
Tª
ºC
NaMk-I
Phase
E
NaMk-P
H
%
200
400
E
H
%
15
0.7
82
Mk
27
1.6
14
Q
95
12
4
GP
23
1.2
89
GP
16
0.7
92
Mk
31
1.8
7
Mk
35
1.6
4
Q
99
15
4
Q
96
14
4
GP
17
0.7
95
Mk
35
1.4
5
700
1000
Phase
GP
GP
75
7.4
94
GP
83
7.6
92
Nepheline
85
9.0
5
Mullite
107
12
8
Mullite
115
10.6
1
o
Fig. 2 Optical images of specimens after the 1000 C exposure: a) NaMk-I, b) NaMk-P
+
believed that in the injection process, migration and accumulation of Na in pores
occurred due to the long and closed curing, leading to the formation of nepheline at
o
about 1000 C accompanied with a volume reduction and resulting voids.
4 Conclusions
Statistical nanoindentation has been successfully used to provide information
about the micro-mechanical properties of different phases and their volume distributions in the geopolymer samples. The manufacturing process was found to
significantly affect the micro-mechanical properties and thermal stability of geopolymer materials produced. A long curing period in closed mould associated with
the injection process appeared to be responsible for a higher degree of reaction
and better mechanical properties of the geopolymeric phase. The long and closed
+
curing, however, seemed to result in higher Na concentration in pore solutions,
which led to formation of nepheline and large porosity at 700 – 1000 ºC. On the
174
I. Beleña and W. Zhu
other hand, the sample produced by direct pouring process using open mould and
short curing period was found to lead to an incomplete reaction of the starting materials and, thus, lower mechanical properties of geopolymers. Generally, the mechanical properties of the geopolymer matrix showed a moderate increase when
the sample was exposed to temperature up to 400ºC, no significant change at 400
o
o
– 700 C, and then an increase more than three times at 1000 C.
Acknowledgments. We are indebted to the Valencian Institut of Small and Medium Size
Industries (IMPIVA) and European Social Fund for the financial support through the Program High Specialization in Industrial Technologies with the project High Specialization in
Composed Nanomaterials with application in the Construction Sector. IMAETA /2005/1.
References
1. Davidovits, J.: Geopolymer, Chemistry and Applications. Institute Géopolymère (Geopolymer Institute) (2008)
2. Hammell, J.A., Balaguru, P.N., Lyon, K.E.: Strength retention of fire resistant aluminosilicate-carbon composites under wet-dry conditions. Compos. Part B-Eng. 31, 107–
111 (2000)
3. Barbosa, V.F.F., McKencie, K.J.D.: Thermal behaviour of inorganic geopolymers and
composites derived from sodium polysialate. Mater. Res. Bull. 38, 319–331 (2003)
4. Barbosa, V.F.F., McKencie, K.J.D.: Synthesis and thermal behaviour of potassium sialate geopolymers. Matter Lett. 57, 1477–1482 (2003)
5. Oliver, W.C., Pharr, G.M.: Measurement of hardness and elastic modulus by instrumented indentation: Advances in understanding and refinements to methodology. J. Mater. Res. 19, 3–20 (2004)
6. Ficher-Cripps, A.C.: Nanoindentation. Springer, Heidelberg (2002)
7. Ulm, F., Vandamme, M., Bobko, C., Ortega, J.A.: Statistical indentation techniques for
hydrated nanocomposites: concrete, bones and shale. J. Am. Ceram. Soc. 90, 2677–
2692 (2007)
8. Zhu, W., Hughes, J., Bicanic, N., Pearce, C.: Micro/nano-scale mapping of mechanical
properties of cement paste and natural rocks by nanoindentation. Mater. Charact. 11,
1189–1198 (2007)
9. Beleña, I., Tendero, M.J.L., Tamayo, E.M., Vie, D.: Estudio y optimización de los
29
27
parámetros de reacción para la obtención de material geopolimérico mediante Si y Al
RMN y DRX. Boletín de la Sociedad Española de Cerámica y Vidrio 43 (2004)