See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/285796227

Chemical additives and their behavior on the refractory castable processing and
properties

Article · January 2009

CITATIONS READS

5 1,908

2 authors:

Ivone Oliveira Victor C Pandolfelli

Universidade do Vale do Paraíba Universidade Federal de São Carlos
30 PUBLICATIONS 397 CITATIONS 677 PUBLICATIONS 10,405 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Victor C Pandolfelli on 11 May 2018.

The user has requested enhancement of the downloaded file.

CHEMICAL ADDITIVES AND THEIR
BEHAVIOR ON THE REFRACTORY CASTABLE
PROCESSING AND PROPERTIES
I. R. Oliveira, V. C. Pandolfelli, Materials Engineering Department - Federal University of São Carlos
Rod. Washington Luiz, km 235, São Carlos, SP – Brazil, [email protected] or
[email protected]

ABSTRACT heterogeneous nucleation on particles surface is favored, the precipi-

Chemical additives may be added to castables for different tation of hydrated phases tend to build up strong connections among
purposes, such as dispersing the matrix particles and controlling the neighboring particles, resulting in the so-called setting phenomenon.
cement hydration reaction. The dispersion state of castable matrices In addition to the increase in the mechanical strength, the formation
has a close relation with their rheological behavior. Additionally, the of hydrates decreases the material permeability inhibiting the drying,
cement hydration reaction reduces the castables permeability, as the difficulty of water vapor migration from the inner part of the
inhibiting the drying process. This paper aims to add some further sample to its surface can increase [4].
knowledge on how different additives influence matrix dispersion The drying of castables can be defined as the loss of free and
and cement hydration by measuring castable properties, such as: chemically adsorbed water during the initial heat up process [5, 6].
flowability, workability, drying behavior and splitting tensile The free water is related to the fraction of added mixing water, which
strength. The results show that some additives not only affect the remains trapped in the pores of the structure after curing. The
particle dispersion (matrix and binder) but also the yielded chemically adsorbed water is associated to the hydrates. The relative
hydrates. Particularly, the polyglycol additive favored both the amounts of free and chemically adsorbed water depend on the amount
dispersion of castables and the water consumption reduction, of the water added during castable mixing, on the binder content and
resulting in C3AH6 and AH3 hydrates, which can be correlated to its nature, as well as on the curing temperature and time.
the tensile strength development. On the other hand, in the presence
of citric acid, this formation of hydrates is hindered due to its In previous papers, the authors have shown that the binder’s
retarding effect via calcium ion complexation. hydration reaction is largely influenced by the presence of additives
[7-9]. The chemical additives can be added to the castables with
INTRODUCTION different purposes, such as: (i) dispersing the matrix particles by
Refractory castables, used as linings and other applications when electrostatic/electrosteric stabilization mechanisms (“dispersants”);
high temperature resistance is needed, mainly consist of a raw mate- (ii) enhancing wetting of the raw material by water (“surfactants”);
rial mixture with coarse (5 mm-100 µm) and fine (100-0.1 µm) and (iii) controlling the hydration process of cement particles (either
aggregates, hydraulic binder and chemical additives (either in water “retarders” or “accelerators”).
solutions or suspensions). Nevertheless, the proper selection of the chemical additives may
The calcium aluminate cement (CAC) has been widely used as a have a strong impact on many characteristics of castable processing,
hydraulic binder in refractory castables. This material accounts including the particle dispersion state, homogeneity, rheological
for the processing of monolithic refractories, giving rise to their properties, setting behavior and water demand. More specifically, the
pre-firing mechanical strength. CAC also presents the following particle dispersion state of castables defines the flowability of these
features: fast setting, low cost and ability to provide castables that materials. On the other hand, their working time is mainly ruled by
could resist aggressive environments [1]. the binder hydration process [10].
The cement particle hydration process involves three different This paper aims to add some further knowledge on how different
steps: (i) dissolution of the anhydrous phases; (ii) nucleation of the additives affect the dispersion of matrix particles and the cement
hydrated phases; and (iii) precipitation of the hydrates. When cement hydration reaction by measuring castables properties, such as: flowa-
particles are in contact with water, calcium aluminate anhydrous bility, workability, drying behavior and splitting tensile strength.
phases dissociate and release Ca2+ and Al(OH)4- ions. The ion MATERIALS AND METHODS
concentration increases rapidly and reaches supersaturation. At this
point, the nucleation and growth of the hydrates occur, followed by The refractory castables used in this study were formulated
their precipitation. As a result of the precipitation process, Ca2+ and according to the Andreasen model [11], with packing coefficient q
Al(OH)4- ions are withdrawn from the solution, which now becomes = 0.21 using the PS Designer software [12]. White fused aluminas
unsaturated and allows further dissociation of the anhydrous with particle size distribution ranging from 4750 µm to 11µm were
phases and a continuous hydrate growth. Thus, a cyclic process of used as aggregates (69 wt%). The matrix comprised calcined alu-
dissolution-precipitation is carried out until most (or all) cementitious minas A17NE (22 wt%) and CT3000SG (3 wt%). Calcium alu-
anhydrous particles exposed to water are consumed [2, 3]. As the minate cements of two different suppliers were used as binders

Refractories Applications and News, Volume 14, Number 1 January/February 2009 9

up to 6 wt%: CA14M (Almatis-US) or Secar 71 (Kerneos-France). The setting behavior was evaluated after applying a constant
The characteristics of these raw materials and binders are present- rotation speed (20 rpm) for the castable and the resulting torque as
ed in Table 1. a function of time was recorded. The time corresponding to the first
Different additives were evaluated as dispersant agents: anhydrous torque slope was named workability [7].
citric acid (Labsynth, 192g/mol, 99.5% purity), diammonium Compositions of castables were also cast into 40 mm x 40 mm
citrate (Sigma Aldrich) and the family of polyglycol polymers cylindrical molds for measurements of splitting tensile strength and
developed by Bayer (FS10, FS30 and FS60). apparent porosity, as well as for the drying tests. Samples were
Aqueous suspensions of the matrix (A17-NE) and binders were cured at 50ºC in an acclimatized chamber (Vötsch, model 20-20) in
prepared to evaluate their dispersion as a function of the additive a moisture-saturated environment (~100% RH) for 24, 48, 72 and
concentration. The content of solids in these suspensions ranged 168 hours.
from 57 vol% (A17-NE) to 40 vol% (CA14M, Secar 71), because The splitting tensile strength was measured according to the
of their different particle size distribution and particle agglomera- ASTM C496-90 standard using universal testing equipment (MTS
tion state. Viscosity measurements as a function of dispersant con- Systems, Model 810), on samples just cured (humid) or dried at
tent were carried out with the help of a rotative rheometer RS300 110ºC for 72 hours. A constant loading rate of 42N/s (1 Pa/min) was
(Thermo Haake, Germany) at a shear rate of 50 s-1. applied. The splitting tensile strength was calculated by:
Additionally, zeta potential measurements were performed for (B)
the A17-NE suspensions in the presence of different additives.
Zeta potential measurements were recorded as a function of pH,
where, σf is the splitting tensile value (MPa); P is the maximum
using a probe connected to an automatic data acquisition device
load (N); h (mm) and d (mm) are height and diameter of samples,
(ESA 9800, Matec Applied Sciences–UK). Aqueous solutions of
respectively.
KOH and HNO3 (2N) were used to adjust the pH.
Apparent porosity was evaluated according to the ASTM C830
The raw materials mixing step was carried out in a castable
standard, using kerosene as the immersion liquid. For this measure-
rheometer [13] at constant rotation (44 rpm). The addition of the
ment the samples were dried at 110ºC for 72 hours.
mixing water was carried out in a continuous way in order to
improve the mixing efficiency and the flowability of castables [14, Drying tests were conducted in a thermogravimetric device
15]. After 60s of dry mixing (to homogenize the raw materials), 75 comprising an electric furnace coupled to an electronic recording
wt% of total water required was added up to the turning point. In system [16, 17]. A thin K-type thermocouple was inserted at a
this condition, the castable behaves as a continuous medium. mid-height of the samples, 1 mm beneath their surface. Results
Thereafter, the remaining water was added keeping the same rota- correlating the temperature of the samples with their mass loss rate
tion up to the end of the mixing step. were attained by applying continuous heating (10ºC/min) from
room temperature up to 800ºC, with a dwell time of 20 minutes
The flowability and workability of the castables were evaluated.
at this latest temperature. Mass loss was analyzed using the W
The flowability test consisted of pouring the castable into a stan-
parameter and the dW/dt (drying rate), calculated following equa-
dard cone with a base diameter of 100 mm (ASTM C-230).
tions C and D, respectively:
Following that, the cone was lifted up to allow the mix to flow and
after 60s the spreading diameter was measured. The diameter per-
centage increased after 60s was taken as the self-flow value (C)
according to the following formula:

(A) (D)

where, Df is the final average spreading diameter after lifting up the

conical mold and Di is the initial diameter (100 mm). where, W represents the cumulative fraction of water released during
heating per the total amount of water initially present in the body; M

10 January/February 2009 Refractories Applications and News, Volume 14, Number 1

is the instantaneous mass during the heating stage, Mo is the initial
mass and Mf is the final mass of the sample.
The drying tests were conducted in green humid samples, except for
those cast with CA14M and FS60 additive, which were also tested
after drying at 110ºC for 24 hours. Additionally, the drying test was
carried out in samples prepared either without a binder in the presence
of AC and FS60 additives or without additive, but containing the
binders CA14M and Secar 71.
RESULTS AND DISCUSSION
Flowability and Setting Behavior
The viscosity of the prepared matrix and binder suspensions
were measured as a function of dispersant addition in order to set
its optimum content. The results are presented in Table 2.
Matrix suspensions were also evaluated regarding the development
Figure 1. Zeta potential profiles as a function of pH for alumina suspensions
of electrical charge on the particle surface in the presence of optimum (A17-NE) containing 0.1 mg/m2 of different additives.
dispersant content (0.1 mg/m2). The zeta potential curves as a
function of pH are presented in Figure 1. Citric acid and presented in Table 3. In general, binder dispersion required higher
diammonium citrate resulted in a higher magnitude of repulsive additive amounts compared to the matrix dispersion, which indicates
forces for the suspensions and consequently lower viscosity values that the castable optimization depends mainly on the binder disper-
(Table 2). sion.
Additionally, the effect of these additives on the dispersion of The FS10 additive was the most efficient, resulting in high
refractory castables was evaluated. It is worth noticing that a self-flow (100-112%) and suitable molding time or workability
refractory castable can be considered as a suspension with high solid (33-34 min) (Figure 2). These castable rheological properties are
concentration. Thus, the evaluation of its rheological characteristics is fundamental in defining their placing technique [11, 18, 19].
of great importance, mainly the matrix dispersion, which controls the Self-flow castables fill molds by the action of their own weight,
flowability behavior [18]. For each evaluated composition, the whereas the less fluid castables might require vibration to flow. The
additive content ranged as shown in Table 2. Optimum dispersant pumping method has grown in technological importance for castable
and water contents resulted in homogeneous castables and are placement mostly because of its feasibility for casting, with a better

Refractories Applications and News, Volume 14, Number 1 January/February 2009 11

 and Al(OH)4- ions, and inhibiting the precipitation of the less soluble
hydrate (richer Ca2+). Hence, the hydrate precipitation delay cannot
explain the workability decrease.
The ionic force increase resulting from cement dissolution with the
matrix dispersion must also be taken into account. Both citric acid
and citrate act on matrix dispersion by an electrostatic stabilization
mechanism. These additives undergo adsorption on a particle
surface promoting repulsion among them due to their ionized
groups (O-, COO-), as shown in Figure 1. Their neutralization by
Ca2+ ions reduces the repulsion among matrix particles, giving rise to
agglomeration. This could justify the castable workability reduction
in the presence of these additives. In addition, in the presence of
citric acid, the castable pH is shifted from a value close to 10 to the
Figure 2. Self-flow values (bars) and workability (lines) for castables containing isoelectric point of the matrix-citric acid system (pH ~ 6), which also
CA14M or Secar 71 in the presence of different dispersant additives. induces workability loss.
cost-effectiveness ratio. During pumping, the castable is submitted to These facts show that binder hydration kinetics itself is not enough
high shearing forces inside metallic pipes. Thereby, castables must to determine the castable workability. Instead, the system as a whole
have both rheological and setting behavior compatible to the process, must be considered.
otherwise the risk of clogging or excessive demolding time might Splitting Tensile Strength and Porosity
take place. Thus, one of the main aspects that affect the refractory
Measurements
castable application is the workability.
Castable workability is mainly defined by the hydration kinetics of The results for castables prepared with CA14M or Secar 71 in the
binders. In the presence of additives, such as FS60 and citric acid the presence of different additives, as a function of curing time, are
self-flow reached its highest values, resulting in the most efficient shown in Figure 3.
dispersion of castables at the expense of workability reduction. The For the FS60 containing castables, the highest splitting tensile
speed up of the binder hydration and setting time of castable can both strength values were attained for both binders. The mechanical
explain this phenomenon. strength increased when the samples were submitted to drying at
The addition of citric acid and other citrates is well known in 110ºC, becoming significantly higher when compared to the strength
the literature for their retarding effect on binder hydration mechanism in the presence of other additives.
[8, 9, 20]. This behavior was confirmed by previous tests where the Good results for the FS60 additive can be mainly related to the
setting time of the binder suspensions was evaluated in the presence efficient dispersion of their particles resulting in a well-packed
of different additives [8, 9]. Among the evaluated additives, citric structure with lower porosity, as well as to the lower water con-
acid and diammonium citrate showed the highest retarding effects [8, tent required for their processing (Table 3). This is in tune with
9]. Their ionizable sites generate anions with great affinity to calcium the apparent porosity results for castables in the presence of this
ions, leading to the decrease of the concentration ratio between Ca2+ additive, as shown in Figure 4.

Figure 3. Splitting tensile strength (humid samples or dried at 110ºC) as a function of curing time for castables prepared with CA14M or Secar 71 cement containing
different dispersant additives.

12 January/February 2009 Refractories Applications and News, Volume 14, Number 1

 The drying of low permeability refractory castables, such as CAC
bonded ones, is carried out in several stages, driven by different
mechanisms of mass and energy transport. The first stage consists of
evaporation of free water below 100ºC. After evaporation, the second
stage led by vaporization takes place. The hydrates decomposition is
carried out in the third step. This stage is characterized by the with-
drawal of the chemical water combined to the hydraulic binder. The
temperature range in which the water withdrawal is carried out is a
characteristic of the hydrate type [16,17].
The hydration process of cement particles results in the pre-
cipitation of hydrated phases of calcium aluminate. In general,
CAH10, C2AH8, C3AH6 and AH3 are the main calcium aluminate
hydrates, where C=CaO, A=Al2O3 and H=H2O. During heating up,
these hydrates lose their combined water (dehydration) at a particu-
Figure 4. Apparent porosity for castables prepared with CA14M or Secar 71 in lar temperature and both can be measured by thermogravimetric
the presence of different additives, cured for 72 hours at 50ºC. analysis (drying tests).
Drying Behavior The hydrate CAH10 (CaO•Al2O3•10H2O) and the alumina gel
lose their water of crystallization at low temperatures (100-130ºC).
The drying behavior for castables prepared with CA14M or Secar The C2AH8 (2CaO•Al2O3•8H2O) presents a decomposition in
71 in the presence of different additives, as a function of curing time, the temperature range of 170-195ºC. The C3AH6
is presented in Figure 5. (3CaO•Al2O3•6H2O) often shows a stepwise dehydration process

Figure 5. Drying rate as a function of sample temperature for castables prepared with CA14M or Secar 71 containing different additives, after curing for 24 and 168
hours at 50ºC.

Refractories Applications and News, Volume 14, Number 1 January/February 2009 13

resulting in a metastable compound, C3AH1.5. After losing 4.5 mol
water molecules (300-360ºC). The remaining water molecules are
released close to 450ºC. AH3 (Al2O3•3H2O) usually decomposes
between 210 and 300ºC, but may be converted to boehmite (AH),
which only dehydrates between 530 and 550ºC [21].
The predominant types of hydrates, which show different struc-
tures, are formed according to the sequence shown in Figure 6.
The hydrate CAH10 is formed at the beginning of the precipitation,
and is the main product when curing is carried out at low tempera-
tures (5-10°C) and with enough water. The hydrate C2AH8 is
formed at temperatures between 22 and 35°C. On the other hand,
C3AH6 is the main hydrate formed at temperatures higher than 35°C
or in the presence of a low water amount. The hydrate AH3 is formed
in all the temperature range and water content, although the
increase of temperature and time can change its structure from
amorphous to crystalline.
Among the different phases of calcium aluminate hydrates, C3AH6
is the most stable and CAH10 the least [22, 23]. For AH3, the stability
increases after changing from amorphous to crystalline structure. For
castable curing conditions where the resulting phases CAH10 and
AH3 gel are favored, the increase of time and temperature renders
their gradual replacing by C3AH6 and crystalline AH3 phases [24-26].
In the drying test carried out (Figure 5) some dehydration
peaks can be observed. However, in the presence of the FS60
additive the peaks are not well defined mainly below 400ºC.
Thus, for the latter, the thermogravimetric tests were repeated using Figure 7. Drying rate as a function of sample temperature for castables prepared
samples dried at 110ºC, as presented in Figure 7. The resulting peaks with 6 wt% CA14M and the FS60 additive, humid or dried at 110ºC, after curing
for 24 and 168 hours at 50ºC.
indicated the presence of hydrates that could be identified by the
decomposition temperatures: C2AH8 (170-195ºC), AH3 (210-300ºC), retarding effect of the AC additive on cement hydration mechanism
C3AH6 (350ºC) and AH (550ºC). by Ca2+ ions complexation, inhibits the formation of this hydrate type.
After analyzing the drying curve results, it can be inferred that Thus, the characteristic dehydration peak height of C3AH6 is higher
the water content present in the castable influences the kind of at 168 h than at 24 h, indicating its formation delay (Figure 5).
hydrate formed, according to the diagram for the CAC hydration The use of the FS60 shows as an advantage a fast development of
reaction (Figure 6). With the increase of water content neces- hydratable phases and a reduction in the water content, which helps
sary for processing the castable (AC>FS10>FS60), the C2AH8 reduce the demolding time of castables.
phase was always present, as indicated by its characteristic
The influence of each additive on the type of hydrate formed can be
dehydration peak close to 200ºC.
better visualized when the drying curves are compared for the same
Conversely, in the presence of the FS60 additive, the lower curing time (Figure 8). Three well-defined regions can be identified:
water content necessary for the castable processing induces the (1) region A, below 200ºC, related to the presence of free-water
C3AH6 formation, after curing at 50ºC. and hydrates with higher water content (C2AH8); (2) region B
In the presence of the AC additive, a reduction of C3AH6 formation (200-400ºC) corresponding to the hydrates C3AH6 and AH3; and (3)
when compared to the FS10 additive can also be observed. The region C (500-600ºC) due to the AH.
Thus, in the presence of AC and FS10 additives, dehydration
occurred in the region A, and the peaks in region C were less noticed
mainly in the presence of AC, after curing for 24 hours. When the
FS60 additive was used, the main formation of hydrates occurred in
regions B and C.
In agreement with the literature [27], the C3AH6 formation (a
calcium rich phase) is followed by the AH3 one due to the high
amount of AlOH4- ions left. This phase, usually a gel, shows a high
capacity to fill interparticle gaps inducing the strength development.
Thus, the formation of C3AH6 in the presence of FS60 explains the
greater amount of AH3 and the consequent higher splitting tensile
strength.
The dehydration of AH3 (210-300ºC) resulted in the AH phase,
increasing the corresponding peak in region C. The more C3AH6 is
Figure 6. General diagram for hydration reaction of CAC [22].

14 January/February 2009 Refractories Applications and News, Volume 14, Number 1

Figure 8. Drying rate as a function of sample temperature for castables prepared with CA14M or Secar 71 containing different additives, after curing for 24 and 168
hours at 50ºC.

formed, the higher the peaks detected in regions B (C3AH6 + AH3)

and C (AH).
Thus, the additives evaluated in this work not only act as dispersing
particles of the matrix and binder, but also define the type of formed
hydrate and influence the castable properties, such as drying behav-
ior and splitting tensile strength. The effect of these additives on
castable high temperature properties is under investigation by the
authors.
The drying tests were also performed for compositions without
additives, as shown in Figure 9. The free-water content withdrawn
during heating was significantly increased (peak ~ 100ºC). This
indicates that the binder hydration process, as well as hydrate forma-
tion depends on dispersion state of its particles. Thus, in the absence
of additives, there is a greater free-water content, which will be
released during drying. Moreover, in this case, the C3AH6 (200-
400ºC) was induced due to the lower water content used for castable
processing (4.5 wt%).
The samples molded without additives presented many defects,
resulting in a more porous structure when compared to those attained
with well-dispersed castables. This structural characteristic seems to
influence the type of hydrate formed, because a denser structure
renders also the precipitation of AH3, as observed in the presence of
FS60.
Another factor that can induce the AH3 formation is the tem-
perature. Drying at 110ºC resulted in a significant increase of split-
ting tensile strength for samples where the FS60 additive was present,
when compared to humid samples (Figure 3).
By comparing the drying curves in the presence of different binders, Figure 9. Drying rate as a function of sample temperature for castables prepared
the following conclusions can be drawn: (i) AH and AH3 hydrates are with CA14M or Secar 71 without additives, after curing for 24 and 168 hours at
50ºC.

Refractories Applications and News, Volume 14, Number 1 January/February 2009 15

formed mainly in the presence of Secar 71 when FS10 is used as an 9. I.R. Oliveira, V.C. Pandolfelli, “Hydration of Binders in the
additive, increasing the tensile strength after drying at 110ºC (Figure Presence of Matrix and Additives”, Cerâmica, 53 [325] 240-248
3), and, (ii) when FS60 is used as an additive, the mechanical strength (2007). (in Portuguese).
attained is higher when compared to Secar 71-FS10 system, regard- 10. I.R. Oliveira, A.R. Studart, F.A. Valenzuela, V.C. Pandolfelli,
“Influence of Additives on Workability of ULC Refractory
less of the binder used.
Castables”, Cerâmica, 49, 11-16 (2003). (in Portuguese).
Thus, FS60 additive can be pointed out as the most efficient among 11. I.R. Oliveira, A.R. Studart, R.G. Pileggi, V.C. Pandolfelli,
the different additives hereby evaluated, resulting in highly dispersed “Dispersion and Particles Packing – Principles and Applications in
castables with high flowability and higher splitting tensile strength. Ceramics Processing”, Making Art Editorial, 195p (2000).
On the other hand, the denser structure attained in the presence of this 12. R.G. Pileggi, “Tools to Study and Develop Refractory Castables”,
additive can inhibit the castable drying, increasing the risk of explo- Ph.D. Thesis, 187p (2001).
sion, depending on the heating rate. 13. R.G. Pileggi, V.C. Pandolfelli, A.E. Paiva, J. Gallo, “Novel
Rheometer for Refractory Castables”, Am. Ceram. Soc. Bull., 79
CONCLUSIONS [1] 54-58 (2000).
The suitable dispersion of a hydraulic binder greatly enhances 14. R.G. Pileggi, A.R. Studart, V.C. Pandolfelli, “How Mixing Affects
refractory castable flowability. the Rheology of Refractory Castables Part I”, Am. Ceram. Soc.
Bull., 80 [6] 27-31 (2001).
Additives, such as citric acid and citrate reduces the workability of 15. R.G. Pileggi, A.R. Studart, V.C. Pandolfelli, “How Mixing Affects
the castable due to their affinity to calcium ions, increasing cement the Rheology of Refractory Castables Part II”, Am. Ceram. Soc.
dissolution and consequently reducing the electrostatic repulsion Bull., 80 [7] 38-42 (2001).
among matrix particles. 16. M. Innocentini, F. Cardoso, M. Akiyoshi, V. Pandolfelli, “Drying
The additive FS60 favors both the dispersion of castables and the Stages During the Heat-Up of high-Alumina, Ultra Low Cement
Refractory Castables”, J. Am. Ceram. Soc., 86 [7] 1146-1148
water consumption reduction, resulting in C3AH6 and AH3 hydrates, (2003).
which can be associated to the splitting tensile strength development.
17. M. Innocentini, F. Cardoso, M. Miranda, V. Pandolfelli,
On the other hand, in the presence of citric acid, the formation of “Vaporization Process and Pressure Buildup During the
these hydrates is hindered due to its retarding effect by calcium ion Dewatering of Dense Refractory Castables”, J. Am. Ceram. Soc.,
complexation. Furthermore, the presence of higher water content in 86 [9] 1500-1503 (2003).
the castable induces the formation of C2AH8 hydrate. The AH3 for- 18. A.R. Studart, W. Zhong, V.C. Pandolfelli, “Rheological Design of
mation is favored when Secar 71 is used as a binder, resulting in a Zero-Cement Self-Flow Castables”, Am. Ceram. Soc. Bull., 78 [5]
greater splitting tensile strength when compared to CA14M. 65-72 (1999).

ACKNOWLEDGMENTS
19. A.R. Studart, R.G. Pileggi, V.C. Pandolfelli, “High-Alumina
Multifunctional Refractory Castables”, Am. Ceram. Soc. Bull., 80
The authors would like to acknowledge Almatis and Kerneos for [11] 34-40 (2001).
supplying the raw materials used in this work, as well as FAPESP and 20. B.R. Currell, R. Grzeskowlak, H.G. Midgley, J.R. Parsonage, “The
CNPq for the financial support and Dr. Hebert Rossetto for the sug- Acceleration and Retardation of Set High Alumina Cement by
gestions. Additives”, Cement and Concrete Research, 7 420-432 (1987).
21. T.D. Robson, “Aluminous Cement in Refractory Castables”, in the
REFERENCES Chemistry of Cements, Vol 2, Academic Press, 1964.
1. T.A. Bier, N.E. Bunt, C. Parr, “Calcium Aluminate Bonded 22. W.E. Lee, W. Vieira, S. Zhang. K.G. Ahari, H. Sarpoolaky, C.
Castables: Their Advantages And Applications”, in: The Latin Parr; “Castable Refractory Concretes”, Int. Mater. Rev., 46 145-
American Association of Refractory Manufacturers Meeting. 167 (2001).
Proceedings Buenos Aires, ALAFAR 75-84 (1996). 23. D. Sorrentino, F. Sorrentino, M. George, “Mechanisms of
2. K.M. Parker, J.H. Sharp, “Refractory Calcium Aluminate Hydration of Calcium Aluminate Cements”, Material Science of
Cements”, Br. Ceram. Trans. J., 81 35-42 (1982). Concrete IV, Edited by J. Skalny and S. Mindess, ACS.
3. C.M. George, “Aspects of Calcium Aluminate Cement Westerville, 41-90 (1995).
Hydration”, Refractories Symposium, The American Ceramic 24. M.R. Nilforoushan, J.H. Sharp, “The Effect of the Time on the
Society, St. Louis 1-21 (1994). Hydration Reactions of Refractory Calcium Aluminate Cement
4. M.D.M. Innocentini, J. Yamamoto, C. Ribeiro, R.G. Pileggi, A.C. Investigated by SEM”, UNITECR ‘01 Proceedings, Cancun
Rizzi, L.R.M Bittencourt, R.P. Rettore, V.C. Pandolfelli, “Effect of 1586-1596 (2001).
Granulometric Distribution in the Refractory Castable Drying”, 25. S.M.B. Watson, J.H. Sharp, “The Effect of Temperature Upon
Cerâmica, 47 [304] 212-218 (2001) (in Portuguese). the Setting Behavior of Refractory Calcium Aluminate
5. J.S. Reed, Drying, Ceramic Age, 87 [7] 81 (1971). Cements”, Cement and Concrete Research, 16 875-884 (1986).
6. G.W. Scherer, “Theory of Drying”, J. Am. Ceram. Soc., 73 [1] 3- 26. C.M.M. Freitas, P.R.G. Brandão, R.P. Rettore, “Hydration
14 (1990). Conditions and Microstruture Development in Calcium
Aluminate Cements’ Paste”, UNITECR ‘93 Proceedings, São
7. I.R. Oliveira, A.R. Studart, F.A. Valenzuela, V.C. Pandolfelli, Paulo 684-696 (1993).
“Setting Behavior of Ultra-Low Cement Refractory Castables in
the Presence of Citrate and Polymethacrylate Salts”, J. Eur. 27. R. Turriziani, “The Calcium Aluminate Hydrates and Related
Ceram. Soc., 23, 2225-2235 (2003). Compounds”, The Chemistry of Cements, Vol. 1, Edited by H.
F. W. Taylor, Academic Press, 233-286 (1964). R AN
8. I.R. Oliveira, J.R. Garcia, V.C. Pandolfelli, “Influence of
Dispersant and Accelerator Additives on Hydration of calcium
Aluminate Cement and Cement-Matrix”, Cerâmica, 52 [323] 184-
193 (2006) (in Portuguese).

16 January/February 2009 Refractories Applications and News, Volume 14, Number 1

View publication stats