Embed Size (px)
Citation preview
Effect of kaolin on the performance of PVA reinforced fiber cement
Beatriz Rodrigues da Silva1,a, Rui Barbosa de Souza1,b and Vanderley Moacyr John1,c
1Av. Prof. Almeida Prado, 83 – São Paulo – SP – Brazil – 05508-070
[email protected], [email protected], [email protected]
Keywords: Fiber cement; kaolin, drying shrinkage, squeeze-flow.
Abstract. The fiber cement industry can use a wide range of materials as fillers. These fillers are
used mostly to reduce costs of the raw materials, but they also can be used to control relevant
properties of the product and to facilitate the operation of the Hatschek machines. It is known that
fillers may also affect mechanical properties as well as drying shrinkage. There is a large experience
on the use of limestone, silica, and fly-ash as fillers. This paper aims at measuring the effect of
kaolin on mechanical properties, drying shrinkage, and rheology of PVA reinforced fibrocement.
Experimental results compare the reference composition (Portland cement, limestone, cellulose, and
PVA fibers) with samples with various amount of kaolin, added as partial replacement of limestone
filler. The samples were produced in laboratory using vacuum filtration, which emulates a Hatcheck
machine. The rheological behavior of fresh mats was assessed by squeeze flow test. Porosity,
drying shrinkage, and MOR were measured by conventional methods. The results show that kaolin
increases the plastic deformation of the composite, making the conformation of the products easy.
This may be attributed to the plate-like shape of its particles, high specific surface area, and surface
charges. The use of kaolin does not change the mechanical properties of the fiber cement.
Introduction
The fiber cement industry can use a wide range of materials with a filling function. It can be
used with filler, limestone, silica, fly ash, and other types of pozzolan. These fillers are used mostly
to reduce costs of the raw materials, but they also can be used to control relevant properties of the
product and to facilitate the operation of the Hatschek process.
It is known that fillers may also affect mechanical properties as well as drying shrinkage. For
example, silica can increase flexure strength, but it can also increase shrinkage [1]. There is a large
experience on the use of limestone, silica, and fly-ash as fillers in cementitious composites.
Kaolin is an ore that can be found in the nature, and its predominant mineral is the kaolinite
(higher amount to 90%). This mineral belongs to the aluminosilicate group.
𝐴𝑙2𝑂3 ∙ 2𝑆𝑖𝑂2 ∙ 2𝐻2𝑂450−700°𝐶→ → 𝐴𝑙2𝑂3 ∙ 2𝑆𝑖𝑂2 + 2𝐻2𝑂 (Eq. 1)
The triclinic crystalline system of the kaolin results in particles with plate geometry [2]. This
quality makes a wide influence on the rheological properties of the fiber cement paste. The shape of
kaolin particles is expressed in terms of the aspect ratio, which is defined as average diameter
divided by thickness [2].
Adding kaolin to the fiber cement can change the parameters of the Hatschek process, such as
the particle retention in the Sieve Cylinder, the compression efficiency of the fresh fiber cement, the
fresh humidity, molding the corrugated sheets, etc. Besides improving the performance of the
products, such as the aspect of corrugated sheets, alteration in hygroscopic movement, decrease of
edge cracking [3], etc.
This paper aims at measuring the effect of kaolin on mechanical properties, drying shrinkage,
and rheology of PVA reinforced fiber cement.
Methodology
Experimental results compare the reference composition (Portland cement, limestone, cellulose,
and PVA fibers) with samples with 7% of kaolin, added as partial replacement of limestone filler.
The samples were produced in laboratory using vacuum filtration, which emulates a Hatcheck
machine.
The ratio of the raw materials used on samples made in the laboratory is representative
compared to fiber cement products industrially produced in Brazil, with 70.0% of cement, 25.1% of
filler, and 4.9% of organic fibers (PVA and cellulose pulp). The samples were produced by the
method suggested by Savastano [4], whose principle is the same as the one involved in the
Hatschek manufacturing process, which is filtering a reactive suspension followed by compression
(compression pressure: 1 MPa) (Fig. 1).
Fig. 1. Laboratory fiber cement production. Sequence: mixture, filtration, and compress.
The rheological behavior of fresh mats was assessed by squeeze flow test. The squeeze flow
test consists in compressing the sample fiber cement and applying a fixed load of 500 N in a
puncture to 2.54 cm of diameter. This test measures the deformation in the compression. 20
measurements for each sample were made (5 measurements for specimen with 200x200 mm).
Fig. 2. Squeeze flow test.
The kaolin effect on cement hydration kinetics was measured by isothermal calorimetry. This
test was made at 23ºC, with cement paste, limestone, kaolin, and water, with a 0.4 water/solids
ratio. This equipment measures the heat released by chemical reactions in cement.
Fig. 3. Isothermal calorimetry.
Porosity was determined by Archimedes’ principle, which determines the dry mass, immersed
mass, and saturated mass of samples.
𝑃 =𝑚𝑠𝑎𝑡 −𝑚𝑑𝑟𝑦
𝑚𝑠𝑎𝑡 −𝑚𝑖𝑚 (Eq. 2)
Where: P – porosity [%]; Msat – saturated mass [g]; Mdry – dry mass [g]; and Mim – immersed
mass[g].
Drying shrinkage was calculated by measuring the length and mass of samples when exposed in
a controlled environment, with relative humidity of 50% and temperature of 23°C.
Fig. 4. Drying shrinkage equipment.
The mechanical performance of the fiber cement was measured by the flexural strength test.
From this test, it was determined the modulus of rupture (MOR) and the limit of proportionality
(LOP), which is the maximum load of the elastic period.
𝑀𝑂𝑅 =𝑃𝑚á𝑥. 𝐿
𝑏. 𝑒² (Eq. 3)
𝐿𝑂𝑃 =𝑃𝑙𝑜𝑝. 𝐿
𝑏. 𝑒² (Eq. 4)
Where: Pmax – maximum test load (N); L – distance between the supports (135 mm); b – width
of the sample [mm]; and thickness of the sample [mm].
The raw materials were characterized by means of laser particle size distribution assays, BET
specific area, and density of He gas pycnometry. Images were made from SEM to characterize the
kaolin.
Results
Materials characterization. The materials used in this study were:
Ordinary Portland Cement (OPC) Portland cement type CPII F-32 (according to the
Brazilian standard NBR 11578 [6])
Limestone filler
Cellulose pulp
PVA fiber
The cement chemical composition used is shown in Erro! Fonte de referência não
encontrada..
Table 1. Chemical composition by XRF [%].
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K2O
OPC 18.7 4.13 2.85 64.5 5.44 2.10 0.11 0.91
Among the materials that make up the cement matrix of fiber cement samples, limestone has a
higher particle size (most frequent diameter of 55 µm). The cement has the most frequent diameter
of 22 µm. The kaolin is the thinnest raw material, with the most frequent diameter of 3.6 µm (Fig.
5).
Fig. 5. Particle size distribution of the raw materials.
Table 2. Real density (γ) and surface area BET (SBET).
Material γ [g/cm³] SBET [m²/g]
Cement 3.154 1.59
Kaolin 2.613 8.91
Limestone 2.730 1.15
In addition to the smaller size particles, kaolin shows a surface area higher than other materials,
with 8.91 m²/g compared to 1.59 e 1.15 m²/g of cement and limestone, respectively (Erro! Fonte
de referência não encontrada.).
The large surface area comes from the geometry of kaolin, presenting plate format. The
micrographs shown in Fig. 6 and Fig. 7 confirm this geometric aspect of kaolin. In Fig. 6, it is
possible to observe agglomerates of kaolin particles having smaller thickness in one of its direction.
Fig. 7 shows a SEM of limestone, when particles are larger and with geometry like cubes.
0
1
2
3
4
5
6
7
8
9
10
0,0001 0,001 0,01 0,1 1
Vo
lum
e [
%]
Particle size [mm]
Cement
Kaolin
Limestone
Fig. 6. SEM images of kaolin.
Fig. 7. SEM images of limestone.
The results of thermogravimetry show that kaolin has 92.36% of kaolinite, because the loss of
volatile (chemically combined) water at around 550°C was 12.89%.
Fig. 8. Thermogravimetry results.
Hydration kinetics. Being an aluminosilicate, the chemical composition of kaolin is compatible
with a pozzolanic material. However, in order to react with cement, it is necessary that the kaolin
not be 100% crystalline. To measure the reactivity of kaolin in the first hours of hydration,
isothermal calorimetry tests were made in different kinds of pastes containing different amounts of
kaolin in their composition (from 0% to 25.2% by mass).
The results shown in Fig. 9 indicate that there was no significant difference in the accumulated
heat after 72 hours of hydration. The kaolin does not react chemically with the cement during the
first hours of hydration. However, the pozzolanic reaction can occur over time, and it is possible
that there be a chemical interaction between the kaolin and hydrated cement matrix.
Fig. 9. Total accumulated heat of the cement hydration.
The only change observed due to the addition of kaolin in the cement paste was a delay in the
start of setting time, or early acceleration period (period of rapid precipitation of C-S-H and
portlandite). At lower levels (up to 7%), the beginning of the acceleration period can take up to 13
minutes more to happen. In most kaolin content tested (25.2%), this time was 49 minutes (Fig. 10).
The precipitation of C-S-H and portlandite occurs when the reactive suspension reaches a
supersaturation level of Ca2+ ions. The addition of kaolin to replace limestone reduces the amount
of calcium suspension, and it also reduces the pH, this way delaying the start of setting time.
100 200 300 400 500 600 700 800 900Temperature /°C
86
88
90
92
94
96
98
100
TG /%
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
DTG /(%/min)
Mass Loss (Marsh): Onset:Mid:Inflection:End:Mass Change:
487.8 °C543.7 °C543.4 °C599.5 °C-12.89 %
Residual Mass: 85.44 % (998.5 °C)
[1]
[1]
0
50
100
150
200
250
300
0 10 20 30 40 50
To
tal a
cc
um
ula
ted
he
at
[J/g
]
Time [hours]
REF (0% kaolin)
2% kaolin
5% kaolin
7% kaolin
25.2% kaolin
Fig. 10. Heat flow in the first 5 hours of cement hydration. Effect of the kaolin content on initial setting time.
Rheology properties. The kaolin has characteristics that strongly influence the rheological
behavior of the pastes. The small particle size, the extensive surface area, and plate geometry make
the pastes require more water to flow. The acting surface forces the adsorption of water between the
particles, facilitating their movement.
A sample of cement with kaolin showed lower resistance load, with more deformation for an
applied load of 500N. This result shows that fiber cement containing kaolin has lower viscosity at
this shear rate level. The viscosity of the paste can be determined by the maximum strain or load
relaxation. Fig. 12 shows the correlation between load relaxation and compressive extension,
furthermore the sample with kaolin showed higher compressive extension and higher load
relaxation.
The dispersion with kaolin was more difficult because this material tends to agglomerate due
to its affinity with water. Therefore, the results with kaolin have shown more variability.
Fig. 11. Squeeze-flow results of the ref.
0
1
2
3
0 1 2 3 4 5
He
at
flo
w [
mW
/g]
Time [hours]
REF (0% kaolin)
2% kaolin
5% kaolin
7% kaolin
25.2% kaolin
0
1
2
3
0% 10% 20% 30%
Init
ial s
ett
ing
tim
e [
ho
urs
]
Kaolin content [%]
0
100
200
300
400
500
600
0 1 2 3 4
Co
mp
res
siv
e lo
ad
[N
]
Compressive extension [mm]
0
100
200
300
400
500
600
0 1 2 3 4
Co
mp
res
siv
e lo
ad
[N
]
Compressive extension [mm]
Fig. 12. Correlation between load relaxation and compressive extension.
Shrinkage. The shrinkage is an important property of fiber cement, especially due to a recurring
industrial problem, which is the differential shrinkage on the edge of the corrugated plates,
generating cracking.
Studies show that the addition of fine materials in the cementitious matrix may increase the
shrinkage [1],[7]. However, the results show that the kaolin did not change the shrinkage of fiber
cement (Fig. 13).
Fig. 13. Drying shrinkage results.
Fig. 14. Shrinkage susceptibility.
FC.ref
FC.kaolin
90
95
100
105
110
115
2 3 4
Lo
ad
re
lax
ati
on
[N
]
Compressive extension [mm]
-3
-2
-1
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Sh
rin
ka
ge
[m
m/m
]
Drying time [days]
REF
Caulim
-3
-2
-1
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Sh
rin
ka
ge
[m
m/m
]
Loss mass [%]
REF
Caulim
The relationship between shrinkage and mass loss during drying indicates susceptibility to
shrinkage of fiber cement, i.e. it shows that for the same mass loss, two samples may have different
shrinkage. Fig. 14 shows that the susceptibility to shrinkage of both samples is similar, indicating
that the kaolin did not alter this property of the fiber cement.
The total shrinkage of the cement depends on the porosity and pore size distribution of the
cement matrix [8]. This is because shrinkage is a function of the amount of water stored in the
material, and function of the pore size distribution, since the shrinkage is caused by capillary
pressure acting on pores smaller than 2 µm.
The total porosity of the cement sample was smaller in the sample with kaolin than in the
reference sample, but this difference is small and within the range of the variation of results, as
shown in Fig. 15.
The kaolin promoted a refinement of the fiber cement pores, with decrease in pore volume
between 2 and 20 µm. However, these changes are not reflected in shrinkage, since the pores of this
size do not suffer from capillary pressure drying.
Fig. 15. Total porosity by Archimedes’ principle.
Fig. 16. Mercury intrusion porosimetry.
Flexural strength. The mechanical strength of the fiber cements was measured by the flexural
strength test. The modulus of rupture (MOR) and the limit of proportionality (LOP) were
determined for this test.
The differences in flexural performance of the reference samples and kaolin were few, always
within the standard deviation of the average range for the MOR, LOP and elastic modulus.
However, it was observed that the kaolin has improved mechanical strength of fiber cement in all
cases (Fig. 17 and Fig. 18).
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
REF Kaolin
To
tal p
oro
sit
y [
%]
0
0,005
0,01
0,015
0,02
0,025
0,01 0,1 1 10 100 1000 10000
Intr
ud
ed
vo
lum
e [
ml/
g]
Pore diameter [µm]
REF
Kaolin
The kaolin increased the rigidity of the matrix, with an increased modulus of elasticity. The
paste strength also increased, indicated by the higher LOP sample containing kaolin. The maximum
resistance indicated by the MOR was also higher in the sample with kaolin, indicating the potential
of improving the mineral addition in fiber cement strength.
Fig. 17. Flexural performance of fiber cement REF (left) and fiber cement with kaolin (right).
Fig. 18. Flexural performance of fiber cement. MOR, LOP, and elastic modulus.
Conclusions
According to the results presented, it can be concluded that the use of kaolin fiber cement
influences its rheological properties positively. It improves the flow conditions, which may be
reflected in an improvement in the appearance of products, ease of conformation (reduction in
longitudinal cracking), better efficiency in compression, and a decrease in the permeability of the
products.
The kaolin did not significantly change the cement hydration kinetics. The increase in the
setting time when using kaolin at 25% is 49 minutes. However, this added amount of kaolin is very
high. In the lower content, benefits to the rheological properties have been achieved, and the change
in hydration kinetics was minimal.
The shrinkage of the cement was not changed, although pore refinement was observed. This
behavior occurred because changes in pore size distribution happened in the upper pores of the
2 µm range, and this pore range did not suffer capillary pressure drying.
The flexural strength increased with the addition of kaolin, due to the increased stiffness of the
cementitious matrix, LOP, and MOR.
0
1
2
3
4
5
6
7
8
9
10
11
12
0 2 4 6 8 10
Fle
xu
ral s
tre
ss
[M
Pa
]
Deflection [mm]
0
1
2
3
4
5
6
7
8
9
10
11
12
0 2 4 6 8 10
Fle
xu
ral s
tre
ss
[M
Pa
]
Deflection [mm]
0
2
4
6
8
10
12
REF Kaolin
Str
es
s (
MP
a)
MOR (MPa) LOP (MPa)
0
2000
4000
6000
8000
10000
12000
REF Kaolin
Ela
sti
c m
od
ulu
s (
MP
a)
E (MPa)
Acknowledgments. The authors would like to thank Imbralit Ltda. and Infibra/Permatex Ltda. for
the raw materials.
References
[1] R. B. de Souza and V. M. John, “The influence of silica fume on the drying shrinkage of fiber
cement reinforced with PVA fibers,” in Proceendings - International inorganic-bonded fiber
composites conference, Aalborg, 2010, pp. 80–89.
[2] A. Psyrollos, J. H. Howe, D. A. Manning, and S. D. Burley, “Geological controls on kaolin
particle shape and consequences for mineral processing,” Clay Minerals, vol. 34, pp. 193–208,
1999.
[3] N. B. Mtsweni, “Effect of natural kaolin on the cracking of corrugated PVA fibrecement
sheets,” in Proceendings - International inorganic-bonded fiber composites conference, Da
Nang - Vietnam, 2014, pp. 143–152.
[4] H. Savastano, P. G. Warden, and R. S. P. Coutts, “Microstructure and mechanical properties of
waste fibre–cement composites,” Cement and Concrete Composites, vol. 27, no. 5, pp. 583–
592, May 2005.
[5] Associação Brasileira de Normas Técnicas, Cimento Portland composto. 1997, p. 5.
[6] I. R. de Oliveira, A. R. Studart, R. G. Pileggi, and V. C. Pandofelli, Dispersão e empacotamento
de partículas - princípios e aplicações em processamento cerâmico. São Paulo: Fazendo Arte
Editorial, 2000.
[7] F. S. Ortega, R. G. Pileggi, P. Sepúlveda, and V. C. Pandofelli, “Influência dos Modelos de
Alfred e de Andreasen Sobre a Microestrutura e Densidade a Verde de Compactos Cerâmicos
Obtidos por Colagem ou Prensagem,” Cerâmica, vol. 43, pp. 183–189, 1997.
[8] J. E. Funk and G. R. Dinger, “Particle packing, part II: review of packing of polydisperse
particle systems,” Interceram., vol. 41, no. 2, pp. 95–97, 1992.
[9] F. de Larrard, Concrete mixture proportioning: a scientific approach. London ; New York: E &
FN Spon, 1999.
[10] J. Hoppe Filho, “Sistemas cimento, cinza volante e cal hidratada: mecanismo de hidratação,
microestrutura e carbonatação de concreto,” Tese (doutorado), Escola Politécnica da
Universidade de São Paulo, São Paulo, 2008.
[11] K. Vance, M. Aguayo, T. Oey, G. Sant, and N. Neithalath, “Hydration and strength
development in ternary portland cement blends containing limestone and fly ash or
metakaolin,” Cement and Concrete Composites, vol. 39, pp. 93–103, May 2013.
[12] H. F. W. Taylor, Cement chemistry, 2nd ed. London: T. Telford, 1997.
[13] R. B. de Souza, “Estudo da retração em fibrocimento reforçado com fibra polimérica,”
Escola Politécnica da Universidade de São Paulo, São Paulo, 2014.
