INSULATING REFRACTORY MATERIALS FROM INORGANIC WASTE RESOURCES by AMANDA JONKER Submitted in partial fulfillment of the requirements for the DOCTOR TECHNOLOGIAE in the Department Chemistry FACULTY OF SCIENCE TSHWANE UNIVERSITY OF TECHNOLOGY Supervisor: Dr MJ van der Merwe Co-Supervisor: Prof RI McCrindle December 2006
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
INSULATING REFRACTORY MATERIALS FROM
INORGANIC WASTE RESOURCES
by
AMANDA JONKER
Submitted in partial fulfillment of the requirements for the
DOCTOR TECHNOLOGIAE
in the Department Chemistry
FACULTY OF SCIENCE
TSHWANE UNIVERSITY OF TECHNOLOGY
Supervisor: Dr MJ van der Merwe Co-Supervisor: Prof RI McCrindle
December 2006
i
I hereby declare that the thesis submitted for the degree D Tech: Ceramics
Technology, at the Tshwane University of Technology, is my own original work
and has not previously been submitted to any other or quoted are indicated and
acknowledged by means of a comprehensive list of references.
“Sonder julle opoffering sou dit nie vir my moontlik gewees het nie.”
iii
ACKNOWLEDGEMENTS The author would like to express gratitude to: Dr MJ van der Merwe, my supervisor and mentor, for her able supervision, criticism and constant readiness to discuss problems during the course of this work and for proofreading the script and sitting through the night with me. Prof RI McCrindle, my co-supervisor, for his efforts, hard work and dedication in finalising this work. Prof JH Potgieter, for motivating me to further my studies. The Department of Chemistry & Physics as well as the Department of Chemical & Metallurgical Engineering, Tshwane University of Technology, for arranging my duties so that I could pursue my studies. The Ceramics Technology division of the Department of Chemistry & Physics, Tshwane University of Technology, for fulltime use of their laboratories and facilities. The National Research Foundation for the financial support to fulfil my studies. Mr MI Lavere for his assistance and hard work in the laboratory (RIP). Miss W Perrins, Cermalab, for her help during the development stages of the project and assistance in testing. All my B. Tech students, for their assistance during the course of this work. Colleagues, family and friends for their critical opinions, aid and patience.
iv
ABSTRACT
The management of inorganic waste produced from diverse forms of industrial
activity remains a major problem in many parts of the world. Typical industrial
inorganic wastes include coal fly-ash, metallurgical slag, phosphogypsum waste
and iron-rich waste.
This investigation focused on the use of coal fly-ash, phosphogypsum and iron-
rich waste as a substitute for natural aluminosilicate raw materials for
2.2.3 SERVICE LIMITING TEMPERATURE The chemical composition, as a basic property of all refractory products,
determines the sintering and melting of heat insulators and, the classification
temperature. As most high-temperature insulating materials consist of silica
(SiO2) and alumina (Al2O3) and the liquidus temperature of the SiO2-Al2O3 system
(Figure 1.1) increases in the high alumina containing section corresponding to
the Al2O3 content, the classification temperature rises with increasing Al2O3-
content in heat insulating materials. Due to the required volume stability, the
16
increased application temperature asks for a higher bulk density with increased
stability and thermal conductivity (Schulle & Schlegel, 1991).
2.2.4 THERMAL CONDUCTIVITY
Thermal conductivity, λ, is defined by Carniglia and Barna (1992) as:
λ (T) = ρ (T).cp(T).a(T) (Eq. 2.1)
where ρ is the bulk density, cp the specific heat, a the thermal diffusivity and T the
temperature. The unit for thermal conductivity λ is W.m-1.K-1.
Unlike the heat capacity, the thermal conductivity of heterogeneous mixtures is
intensely sensitive to variations in microstructure. The governing micro structural
features being intimately dependant on processing and thus largely uncoupled
from composition, there is no reliable “rule of mixtures” for thermal conductivity.
Figure 2.1: Thermal conductivity of insulating fire brick and insulating castables (Carniglia & Barna, 1992). STL indicating the Service Temperature limit (in °F) of the Insulating Fire Brick (IFB)
17
The variation of the thermal conductivity with average body temperature for
insulating fire bricks and insulating castables is consolidated in Figure 2.1.
Internal heat transportation, and with it heat insulation, in high-temperature
insulating materials, are decisively influenced by the structural composition and
the temperature. The effectiveness of the influenced temperature is also
controlled by the structure. Consequently, the structural composition plays a
dominating part. As emphasised before, high-temperature insulating materials
represent heterogeneous, porous multiple phase bodies. These materials
facilitate extensive internal heat transportation by means of thermal conduction
and heat radiation, which can be summed up as an effective thermal conductivity:
• The porosity, or bulk density, has to be adapted to the temperature of the
application, or the temperature gradient, intended to be applied. The
porosity required for a minimum effective thermal conductivity decreases
with increasing temperature of application (Schulle & Schlegel, 1991).
• Porosity exerts the main influence on the effective thermal conductivity.
• In cases of pure heat conduction the gas-filled pores have a small role to
play, the solid matter structures a decisive one.
• The effective thermal conductivity depends on the thermal conductivity of
the pore-free, solid phase.
• The type of pore gas and the gas pressure influence the thermal
conductivity.
• The pores should be as small as possible and efforts should be made to
provide micro-porosity.
• The microstructure of the solid matter should consist of loosely packed
crystal structures and complicated crystal lattices with little symmetry, high
defect density, as well as a substantial poly- or micro crystallinity.
• The microstructure of the solid matter should show little transmission and
a high degree of absorption in the infrared wave range.
• Cracks and coarse pores more than 5 mm have to be avoided.
18
• The overall structure should not allow gas permeability or at least at only
on a small scale.
2.2.5 SHRINKAGE
The shrinkage behaviour of an insulating material is used for evaluating its
maximum possible temperature of application. For this reason non-reversible
length modification is measured over a long period of time at constant
temperatures, the material being heated up on one or all sides in an oxidising
atmosphere without corrosive influences. The classification temperature or the
limit of application temperature corresponds to the temperature which allows a
maximum admissible amount of linear shrinkage. Most countries have
established different shrinkage standards. For refractory lightweight bricks and
concretes there are shrinkages of 1 to 2 % and for refractory fibres 2 to 5 %,
sometimes even up to 7 %. The isothermal heating time, required for thermal
treatment, also fluctuates between 4 and 24 hours (Schulle & Schlegel. 1991).
A typical refractory is based on a mixture of low shrinkage clays with a small
addition of plastic clays, for example ball clay, to ease shaping during
manufacture and impart high green strength before firing (Hancock, 1988).
2.2.6 STRENGTH Kruger (1996) reported the development of castable refractories from coal fly-ash
and cenospheres which have physical and chemical properties that are inherently
beneficial for the manufacture of insulating refractories. Their use imparts
excellent flow properties to the product, thus enhancing the placeability of
monolithic linings. This phenomenon has been ascribed to the lubricating (ball-
bearing) effect of the spherical particles. Insulating refractories based on coal fly-
ash exhibit remarkable strength to density ratios, excellent thermal shock
resistance and an improved ratio of thermal conductivity to bulk density. Most
19
importantly, they are far more cost-effective than competitive products. In
general, the higher the proportion of cenospheres in the product, the better will
be the insulation efficiency and the lower the density. Compressive strength is,
however, slightly lower at higher cenospheres content. The maximum service
temperature of approximately 1250 to 1300 ºC does restrict the use of
cenospheres and coal fly-ash to heat insulating or lower-temperature refractories.
Careful selection of the particle size distribution of the coal fly-ash or
cenospheres ensures optimum particle packing and enables the manufacture of
low-shrinkage refractories (Kruger, 1996).
The need for energy conservation necessitates insulating refractories with
improved performance. The incorporation of cenospheres as part of the
formulation has enabled the manufacture of products (Cenref) that have lower
thermal conductivity and greater strength, which are lighter than the conventional
Moler bricks widely used in industry. A cenosphere refractory can out-perform
competitive products. Besides its superior insulation, its low apparent porosity is
the most significant advantage. This is ascribed to the fact that the cenosphere
refractory consists of isolated spheres lightly fused together; whereas other types
of insulating refractories have interconnecting micro channels. Heat diffusion is
more efficient along these micro channels than across the isolated air within the
spheres. The inability of liquids to penetrate the monolithic cenosphere matrix
also gives these refractories superior acid resistant properties. Service
temperatures of 1300 ºC have been achieved and formulations have been
developed that, at elevated temperatures, provide superior insulation to ceramic
fibre. Due to their excellent in-service performance, domestically developed coal
fly-ash and cenosphere refractories are gaining popularity (Kruger, 1996).
20
2.3 FIREBRICK REFRACTORIES AND THERMAL INSULATION The group of aluminium silicate lightweight refractory bricks (fireclay and mullite
bricks) is the most important and common group of lightweight refractories.
(Hancock, 1988). Raw materials based on Al2O3, SiO2 and sometimes CaO are
used to produce these bricks. Raw materials such as clays, kaolin, fireclay,
sillimanite, andalusite, kyanite, mullite, alumina, alumina hydrate and corundum
are used as a source of alumina (Figure 1.1). In addition to the granulated fine-
grained raw materials, coarse-grained and porous raw materials are also used.
These include lightweight fireclay and hollow spheres (balls) consisting of
corundum or mullite. The “burnout” process is applied most often to the
production of lightweight refractory bricks. Fine saw dust, petroleum coke, lignite
abrasion; fine waste products of cellulose and paperboard (carton) are utilised as
organic materials to be burnt out. Burnout materials with low ash content are
required in order to prevent negative effects on the hot properties of the
refractory materials.
The foam process is a further method of production to achieve high porosity
refractory materials. Special soaps, saponins and sulfonates are used to make
stable foams (Ferguson, 1982). The slurry for the ceramic body is often made
separately from the foam emulsion. Foam and slurry are homogenised in an
intensive mixer. By the controlled mixing of foam and slurry the required bulk
density is adjusted.
Lightweight, low density and high strength refractory bricks can be produced by
mixing in evaporating substances (naphthalene), which have distinctive
differences in their properties when compared with other bricks. Very fine pores
guarantee that high dimensional accuracy of lightweight refractory products is
achieved by casting, centrifuging or pressing (Hancock, 1988). During casting,
the perforated metal moulds (forms) are lined with filter paper before being filled.
Sulphite liquor, gypsum or concrete can be added in order to strengthen the
21
mixture and to speed up the setting. The centrifuging process of large blocks is
very efficient and ensures excellent dimensional stability. Plastic, semi-dry and
dry mixes are shaped by corresponding presses (extrusion, hydraulic or
mechanical presses). The bricks, unfinished cylindrical pieces or blanks, are fired
in chamber furnaces, bogie hearth furnaces or tunnel kilns. The firing
temperature corresponds approximately to the classification temperature
indicated by the producers. Due to high drying and firing shrinkage, cutting or
grinding is necessary for most brick qualities in order to obtain the standard
shapes. Hand forming, vibration or moulding processes produce bricks which are
complicated in shape (Hancock, 1988).
Otero et al. (2004) reported on the preparation of thermal insulating firebricks
from coal fly-ash. Due to its morphological characteristics, physicochemical
properties and pozzolanic activity, coal fly-ash has potential for use in the
production of refractory insulating bricks in combination with clays, a binder
(sodium silicate) and a foaming agent (50 % hydrogen peroxide). The bricks
obtained exhibit the appropriate characteristics of mechanical resistance, porosity
and thermal conductivity.
Vilches et al. (2003) underlined the use of coal fly-ash and titanium waste in
thermal insulation and fireproof applications. Plates were prepared from a mixture
of coal fly-ash (>50 %) and titanium waste (>35 %). Exfoliated vermiculite
(<10 %) was added to make the material more porous and to reduce the density.
The materials produced exhibit high porosity, with average pore diameters
between 0.5 and 10 µm, an average density of 0.74 g.cm-3, and compressive
strength of approximately 0.31 MPa. Differential thermal analysis (DTA) results
showed that the material is stable at high temperatures (>800 ºC).
Refractories are only the start of yet another field of application for coal fly-ash
and its derivatives. Although volumes used are currently modest, these are
bound to increase as the refractory, and more especially the user industries,
22
realise the benefits that can be achieved. Development is continuing on these
materials and the limits have not yet been reached. More products based on
coal fly-ash and cenospheres will soon be seen with even lower thermal
conductivities (Kruger, 1996). Cenospheres are essentially thin-walled glass
spheres with a relative density of less than 1.0. They float on water and are
recovered from the surface of ash disposal ponds and are of similar chemical
composition to fly ash. Fly ash will be discussed in detail in Section 2.6.
2.4 CONVENTIONAL SILICATE-BONDED REFRACTORIES VERSUS GEOPOLYMERS
Previously silicate-bonded materials have been used in refractories. However,
recent research projects on inorganic silicate materials have evolved a new
product called a geopolymer, which can incorporate large amounts of coal fly-ash
in its formulation.
A geopolymer is an inorganic aluminosilicate, synthesised from predominantly
silicon and aluminium materials of geological origin, or by-products such as coal
fly-ash and granulated blast furnace slag (Cheng & Chiu, 2003).
Geopolymers are versatile materials which can form composites with almost any
material, hence providing the possibility of property amelioration in diverse
applications, such as refractory, thermal insulation, fire resistance, etc., by careful
addition of selected materials. Davidovits (1991) pointed out that physical
properties, such as fusion temperature and coefficient of thermal expansion, are
a function of the Si:Al ratio.
Barbosa and Mackenzie (2003a; 2003b) investigated the thermal behaviour of
inorganic geopolymers derived from sodium and potassium polysialate, with
different inorganic fillers and found that, in general, properly cured potassium
polysialate geopolymer showed little sign of shrinkage and melting up to 1400 ºC.
23
Crystalline phases, leucite (KAlSi2O6) and kalsilite (KAlSiO4), form at
approximately 1000 ºC. Silica-rich geopolymers such as potassium polysialate-
siloxo materials are friable above 1200 ºC. Properly cured sodium-based
geopolymers have a melting point around 1300 ºC.
2.5 PRODUCTION OF FOAM GEOPOLYMERS FROM WASTE MATERIALS
Recycling waste materials would aid in the protection of the environment. When
the properties of waste products are such that it is possible to use them for high
added value applications, these products stand a better change of competing
than products made from primary materials.
Coal fly-ash, iron-rich wastes and ball clay have chemical and physical properties
that, in principle, make them suitable for recycling as geopolymeric materials.
The remarkable achievements made through geosynthesis and
geopolymerisation include the production of mineral polymers termed
geopolymers. These inorganic polymeric new materials can polycondense just
like organic polymers, at temperatures lower than 100 °C (Hardjito et al., 2004b).
Historically (Davidovits, 1991) geopolymerisation involves chemical reactions of
aluminosilicate oxides (Al3+ in the fourfold coordination) with alkali polysilicates
yielding polymeric Si-O-Al-O- bonds. The amorphous to semi-crystalline three
dimensional silico-aluminate structures are of the poly (sialate) type (-Si-O-Al-O-),
the poly (sialate-siloxo) type (-Si-O-Al-O-Si-O) and the poly (sialate-disiloxo) type
(Si-O-Al-O-Si-O-Si-O-). Geopolymeric compounds involved in materials
developed for industrial applications are either crystalline or non-crystalline
(amophorous or glassy structures), whereas, several geopolymeric materials of
practical interest are non-crystalline. This viewpoint has been debated (Swaddle,
2001; Provis et al., 2005).
24
These new generation of materials, when applied in the pure form, reinforced or
with fillers, can be used for storing toxic chemicals or radioactive wastes,
manufacturing of special concretes, moulds for moulding thermoplastics and in
making tooling in the aluminium alloy foundries and metallurgy.
High temperature techniques are no longer necessary to obtain materials that are
ceramic-like in their structure and properties. Geopolymers can polycondense
just like organic polymers at temperatures lower than 100 °C. As a result,
geopolymeric materials are easy to make. Their physical properties make them
viable alternatives for many conventional cements and plastics. Their synthesis
at low temperatures with no CO2 emissions is energy-efficient and more
environmentally friendly than many older materials (Van Jaarsveld, van Deventer
& Lukey, 2003).
The polycondensation potential of geopolymers is much higher than that of
cement-based materials. Thus, geopolymer materials possess many
advantageous properties such as mechanical properties, unique high-
temperature (1200 °C) properties, long-term durability, easily recycled, an
adjustable coefficient of thermal expansion, heavy metal ion-fixation and acid
resistance. It is also a “Green Material” because of its low manufacturing energy
consumption and low waste gas emission. The chemical bonds of Si-O and Al-O
are among the most stable covalent bonds in nature. Consequently,
geopolymers are considered as one of the candidates to solve the conflict of
social development against environmental pollution as they can be utilised in the
fields of fire resistance, nuclear wastes solidification, hazardous wastes disposal,
binder, fast reparation, decoration, intelligent material and construction
(Davidovits, 1991; Van Jaarsveld, van Deventer & Lukey, 2003).
Portland cement production is under review due to the high levels of carbon
dioxide released to the atmosphere. Geopolymer concrete is a new material that
25
does not need the presence of Portland cement as a binder. Instead, low-cost
available materials such as coal fly-ash, that are rich in Si and Al, are used and
activated by alkaline liquids to produce the binder. This also has a positive
effect on the environment (Hardjito et al., 2004a).
Since 1972, Davidovits has been developing a kind of mineral polymer material
with the structure of a three dimensional (3D) cross-linked polysialate chain
(-(Si-O)z-Al-O-) which resulted from the hydrolysation and polycondensation
reactions of natural minerals or industrial aluminosilicate wastes such as clays,
slag, coal fly-ash and pozzolan with alkaline activators below 150 oC. This
“inorganic polymer” material was first named “Polysialate” in 1976 (Zhang, Gong
& Lu, 2004). Nine years later, Davidovits coined another term “geopolymer”, in
his US Patent, to represent this family of inorganic polymers. The term
“geopolymer” has been wildly accepted (Davidovits, Davidovics & Davidovits,
1994; Zhang, Gong & Lu, 2004).
A two-step mechanism for the geopolymer reaction was proposed. The first step
can be named “activation step” including the dissolution of starting materials and
the formation of orthosialate acid in a high pH, basic solution. The second step
concerns mainly the further polycondensation between orthosialate acid and
surface silanol groups and the formation of the 3D-cross-linked polysialate
structure, which can be called the “polycondensation step” (Zhang, Gong & Lu,
2004).
2.5.1. GEOPOLYMER CHEMISTRY
Geopolymers are chemically designed as polysialates. Sialate is an abbreviation
for silicon-oxo-aluminate. The sialate network consists of SiO4 and AlO4 -
tetrahedra linked in an alternating sequence by sharing all of the interstitial
oxygens. Positive ions (Na+, K+, Li+, Ca2+, Ba2+, NH4+ and H3O+) must be present
26
in the framework cavities to balance the negative charge of Al3+ in four fold
coordination. Polysialate has the empirical formula:
Mn[(SiO2)z.AlO2]n·wH2O
where: M is a cation, usually an alkali, n is a degree of polycondensation, w ≤ 3
and z is 1, 2 or 3 (Comerie & Kriven, 2003).
Polysialates are chain and ring polymers with Si4+ and Al3+ in four fold
coordination with oxygen, and are amorphous to semi-crystalline. Apart from
poly-sialate (-Si-O-Al-O-), poly-sialate siloxo (-Si-O-Al-O-Si-O-) and poly-sialate-
disiloxo (-Si-O-Al-O-Si-O-Si-O) chemical groupings are also possible structural
units for geopolymers, when the amount of silicate reactant increases in the
reaction system (Comerie & Kriven, 2003).
Geopolymerisation is exothermic and is given schematically in Figure 2.5. It is
assumed that the reactions are carried out through oligomers (dimers or trimers)
that provide the actual unit structure of the three dimensional, macromolecular
edifices. When geopolymeric polymerisation is carried out at ambient
temperature, amorphous or semi-crystalline structures are formed. However,
when the geopolymers are synthesised at hydrothermal setting and hardening
temperatures, in the 150 oC to 180 oC range, the geopolymeric products are
crystalline in structure. The coordination of Si and Al in geopolymers detected by
nuclear magnetic resonance (NMR) is four fold and the X-ray diffraction of
geopolymeric binder is amorphous with no crystalline peak detectable. The
difference between a geopolymeric binder and a geopolymeric product is that the
geopolymeric binder is synthesised from a precursor such as 2SiO2.Al2O3
(calcined kaolinite), at ambient temperature. However, geopolymeric products or
commercial products are different from the binder, because other materials or
metals are involved in the system as an aggregate or reinforcement, such as for
example, sand, SiC, and carbon fiber (Comerie & Kriven, 2003).
3.1.1 DETERMINATION OF THE CHEMICAL COMPOSITION OF THE INORGANIC MATERIALS
The chemical composition of the inorganic minerals was determined by means of
the X-Ray fluorescence analysis (XRF) on an ARL9400XP+ spectrometer and the
results are listed in Table 4.1. The samples were ground to <75 µm in a tungsten
carbide milling vessel, roasted at 1000 °C to determine the Loss on Ignition value
and after adding 1 g sample to 6 g Li2B4O7 fused into a glass bead.
48
3.1.2 SAMPLE PREPARATION OF INORGANIC MATERIALS
The waste materials listed in Table 3.1 were tested for shrinkage, strength and
water absorption.
The sample mixture for each inorganic material listed in Table 3.1 comprised 80 %
of an inorganic material and 20 % ball clay with enough water to achieve an
extrudable mixture.
The mixture was extruded using a hand extruder with a round orifice of about
13 mm. Eighty samples were cut to a length of 150 mm each. Two indent marks,
100 mm apart were made in each of the rods. All the samples were air dried where
after it were dried overnight at 110 °C in a laboratory drying oven. Twenty samples
were fired at each of the following temperatures: 800; 850 and 900 °C in a
laboratory kiln. The firing rate was 2.5 °C.min-1 with a soaking time of 2 hours and
left to cool without any forced cooling.
3.1.3 SHRINKAGE OF INORGANIC MATERIALS
The shrinkage of all the samples was determined using the standard test method
(Jonker, Maree & Van der Merwe, 1998) after drying at 110 °C (green shrinkage)
and after firing (fired shrinkage). The total shrinkage, which is the shrinkage from
the wet state to the fired state, was also calculated. The raw data of the results are
listed in Appendix A. A summary of the results are listed in Table 4.2 and
graphically represented in Figure 4.1.
3.1.4 STRENGTH OF INORGANIC MATERIALS
The twenty dried samples were tested for green strength to determine if the product
would be strong enough before firing to withstand handling.
49
The strength of all the fired samples was determined. The standard test method for
strength determination was used (Jonker, Maree & Van der Merwe, 1998)
employing a Lloyds 200 three-point flexion bend test apparatus with the distance
between the knife-edges at 100 mm. The raw data of the results are listed in
Appendix A. The results are summarised in Table 4.3 and graphically represented
in Figure 4.2.
3.1.5 WATER ABSORPTION OF INORGANIC MATERIALS
After testing the strength the water absorption was determined on the fired extruded
samples. The samples were boiled for three hours in water to give an indication of
the degree of vitrification and porosity. The standard test method was used (Jonker,
Maree & Van der Merwe, 1998). The raw data of the results are listed in Appendix
A. The results are summarised in Table 4.4 and presented graphically in Figure 4.3.
3.2 INTRODUCTION TO DEVELOPING A CERAMIC BODY MIXTURE FROM WASTE MATERIALS
The selected inorganic waste materials were combined and the mix investigated for
refractory bodies conforming to set standards. Refractory body development will
lead to the investigation of traditional methods for the manufacture of lightweight
insulating refractory materials. Due to the lack of plasticity of the mixtures, casting
was the chosen forming method.
3.2.1 SAMPLE PREPARATION OF CERAMIC BODY MIXTURES Guided by the properties of the tested raw materials three mixtures were formulated
as tabulated in Table 3.2. Ball clay was added to increase the strength and
plasticity of the mixture, coal fly ash was utilized as the filler and major source of
50
waste material while the iron rich waste and phosphogypsum were added as fluxing
aids and waste utilisation.
Table 3.2: Body mixtures for ceramic bodies
Material FCB (%)
FGI (%)
FPI (%)
Coal fly-ash 80 80 80
Overburden bentonite 5 - 5
Ball clay 15 15 -
Iron-rich waste - 5 10
Phospho-gypsum - - 5
To each of the three extruded mixtures was added:
• 30% H2O
To each of the three cast mixtures was added:
• 40 % H2O and
• Alcosphere™ deflocculant a sodium polyacrylate.
3.2.2 PARTICLE SIZE DISTRIBUTION OF CERAMIC BODY MIXTURES
The particle size distributions of the mixes were conducted on the Malvern
Mastersizer according to the standard test method (Jonker, Maree & Van der
Merwe, 1998). The results are represented by Figure 4.4. 3.2.3 FLOW PROPERTIES OF CERAMIC BODY MIXTURES The flow properties (fluidity and thixothopy) on all three mixes, with varying amounts
of deflocculant, were determined using the Torsion viscometer to obtain the best
amount of deflocculant to be used for casting (Jonker, Maree & Van der Merwe,
51
1998). This determined the amount Alcosphere™ that was used in the mix for
casting the samples. A summary of the results are represented in Table 4.5 and
figures 4.5 and 4.6.
3.2.4 EXTRUSION AND CASTING OF CERAMIC BODY MIXTURES For each mix 100 samples were cast and 20 samples per mix were also formed into
briquettes with a hand extruder without a de-airing chamber. All the samples were
marked and left to dry in air for approximately a week, after which they were dried at
110 °C in a laboratory drier overnight.
3.2.5 FIRING OF CERAMIC BODY MIXTURES
Twenty prepared samples were fired at each of the following temperatures: 900,
950, 1000 and 1050 °C. The heating rate employed was 2 °C.min-1 with a soak
period of 2 hours, and then cooled naturally.
The physical ceramic properties (shrinkage, strength and water absorption) were
determined according to the same methods mentioned in section 3.1.3 to 3.1.5.
The raw data is listed in Appendix B and summarised in Table 4.6 to 4.9 and
represented graphically in Figures 4.7 to 4.9. Additional tests as listed below, were
also conducted for these mixtures and their samples.
• The wax method (Jonker, Maree & Van der Merwe, 1998) to determine the bulk
density of the final products. The raw data is listed in Appendix B and the results
are summarised in Table 4.7 to 4.9.
• Particle size distribution as obtained by the Malvern Master sizer and are
graphically presented in Figure 4.4.
52
3.3 INTRODUCTION TO THE DETERMINATION OF REFRACTORY PROPERTIES OF POROUS CERAMIC BODIES FROM INORGANIC WASTE MATERIALS
The scope of this section is to determine the refractory properties of different
ceramic mixtures that include a specific organic material. The organic material is
introduced to achieve higher porosities in the mixtures. Higher porosities will result
in lower bulk densities and lower thermal conductivities.
3.3.1 PROCEDURE
The prepared mixtures for porous refractory materials are listed in Table 3.3.
Table 3.3: Body mixtures for porous refractory materials
Material FBO (%)
FBI (%)
FGI (%)
Coal fly-ash 80 80 80
Overburden bentonite 5 - 5
Ball clay 15 15 -
Iron-rich waste - 5 10
Phospho-gypsum - - 5
• One kilogram samples of the body mixture were hand mixed with 30 - 40 % water
and two grades of polystyrene, 80% small beads (0.4 – 0.7 mm) and 60 % large
beads (0.7 – 1.2 mm), to avoid the polystyrene beads being squashed.
• The resulting mixes were separately cast into steel moulds that had been greased
(for easier removal of set samples).
• The moulds were vibrated by hand to prevent separation of the lighter polystyrene
from the rest.
53
• The samples were allowed to set for 24 hours in the steel moulds whereafter it was
removed from the mould and left to air-dry for 24 hours.
• The samples were then dried in a laboratory drier at 110 °C for another 24 hours.
• The dried samples were fired in a laboratory electric furnace at a heating rate of
80 °C per hour to 1100 °C and then soaked for 1 hour. The furnace was allowed
to cool to room temperature without any forced cooling.
The following tests were conducted on the porous refractory mixes and/or samples
to determine their refractory properties by employing the following methods:
• Chemical analyses were determined by XRF as described in section 3.1.1. The
results are represented in table 4.10.
• Mineralogical analysis was done by XRD on the fired samples. The analyses
were done on a Siemens D500 X-ray diffractometer (Bruker AXS, Germany)
using the Cu-Kα line and 2θ angle from 10 to 70 °. The results are represented
in Figure 4.10 to 4.13.
• Ash fusion temperature tests in an oxidising atmosphere to 1550 °C were done
on the FBI, FBO and FGI green mixes to determine the melting temperature of
each mix according to the method described in ASTM D 1857. A pyramid of the
test material is used to determine the softening and other temperatures. The
results are represented in Table 4.11 and Figure 4.14.
• Apparent porosities, of the FBI (large and small beads) and FBO (large and small
beads) mixes, were determined according to the method described in ISO 5016
and ISO 5017. The mass of the dried test piece of a specified geometrical form
is determined by weighing and the dimensions are measured. The results are
represented in Table 4.12 and Figure 4.15.
• Bulk densities of the FBI (large and small beads) and FBO (large and small
beads) mixes, were determined according to the method described in ISO 5016
and ISO 5017. The mass of the dried test piece of a specified geometrical form is
54
determined by weighing and the dimensions are measured. The results are
represented in Table 4.12 and Figure 4.16.
• Apparent relative densities of the FBI (large and small beads) and FBO (large
and small beads) mixes were determined, according to the method described in
ISO 5016 and ISO 5017. The mass of the dried test piece of a specified
geometrical form is determined by weighing and the dimensions are measured.
The results are represented in Table 4.12 and Figure 4.16.
• Cold crushing strength (CCS), was determined according to the method
described in ASTM C133-84 on green and fired samples. The test piece of
specified surface area is compressed with an increasing pressure until
disintegration. The results are represented in Table 4.13 and Figure 4.16.
• Thermal conductivities of the FBI (large and small beads) and FBO (large and
small beads) mixes were determined according to the method described in
ASTM C 201-86 in oxidising atmosphere by Mittalsteel Research Laboratories.
The results are represented in Table 4.14 and Figure 4.17. 3.4 INTRODUCTION TO THE PRODUCTION OF INSULATING
GEOPOLYMERS FROM WASTE MATERIALS The traditional manufacturing method used to produce insulating materials poses
restrictions to the industrialisation of this project. Geopolymerisation might possibly
be the answer to this problem.
3.4.1 PROCEDURE Seven mixtures were formulated as listed in Table 3.4 and to each mixture the
following was added:
• 15 ml 55% sodium silicate (Na2SiO3)
• 10 g sodium hydroxide
• 10 g meta-kaolin
55
Table 3.4: Composition of mixtures for sample geopolymer preparation.
Mixture Material
1 2 3 4 5 6 7
Coal fly-ash (g) 40 30 25 50 50 50
Ball clay (g) 10 20 25 50
Al (g) 0.025 0.005
H2O (ml) 10 10 10 10 15 10 10
The sodium hydroxide was dissolved in the water and the solution was cooled to
room temperature where after sodium silicate was added followed by the addition of
meta-kaolin. The suspension was stirred for 5 minutes. While hand stirring the
mixture the inorganic oxide materials were added to the mix. In mixtures 6 and 7,
aluminium powder (particle size: 40 µm) was added last as a porogen. The
geopolymeric sample was then cast into plastic moulds and left to dry at room
temperature for ± 24 hours.
3.4.2 PHYSICAL PROPERTIES OF INSULATING GEOPOLYMERS The physical properties of the cast geopolymeric mixtures samples were determined
as the mean value of 3 samples.
• The percentage shrinkage was determined after drying at room temperature
(green shrinkage) using the standard test method as described by Jonker,
Maree and Van der Merwe (1998). The results are presented in Figure 4.18.
• The bulk densities of the geopolymeric samples were determined according
to ISO 5016 and ISO 5017. The results are presented in Figure 4.19.
56
• Cold crushing strength (CCS) was determined according to ASTM C133-84.
The results are presented in Figure 4.20.
• The true volume of geopolymeric material was calculated using a stereo
pycnometer according to the standard method described in ISO 5016 and
ISO 5017. The results are represented in Figure 4.21.
• Ash fusion temperature test in oxidising atmosphere to 1550 °C insulating
geopolymeric samples were done to determine the melting temperature of
each mix according to ASTM D 1857. The results are represented in Table
4.15 and Figure 4.22.
• The thermophysical properties were outsourced to the Applications
Laboratory, Thermophysical Properties Section in Germany and conducted
by A. Lindemann and J Blumm. The thermal diffusivity was measured using
a Netzsch model 457 MicroFlash laser flash diffusivity apparatus in a
dynamic argon atmosphere at a flow rate of ~ 100 ml.min-1, between room
temperature and 1100 °C. The specific heat measurements were conducted
using a Netzsch model DSC404C Pegasus differential scanning calorimeter.
The measurements were carried out in a dynamic argon atmosphere (gas
flow rate: 50 ml.min-1). The system was equipped with a temperature-
calibrated DSC-Cp type S sensor. Platinum crucibles were employed for the
test. The sample was heated between room temperature and 1100 °C at a
heating rate of 20 K.min-1. The results are illustrated in Figure 4.23 and
tabulated in Table 4.16.
57
CHAPTER 4
RESULTS AND DISCUSSION
4.1 RESULTS OF PHYSICAL TESTS ON THE INORGANIC MATERIALS
4.1.1 CHEMICAL COMPOSITION OF INORGANIC MATERIALS
The chemical composition of the inorganic minerals was determined to compare the
oxide ratios present, which indicate the mineral phases that may form during firing.
The results are listed in Table 4.1. The accuracy of the XRF analysis is 1 %.
Table 4.1: Chemical composition of the inorganic materials % Lethabo
Traditionally manufactured porous ceramic material
Specification for porous insulating refractories
Comments on compliance to specification
Total linear shrinkage (%) 1.32 4 1 to 7 GPIR the
best
Strength (MPa) 13.84 7.89 0.31 GPIR the best
Density (g.cm-3) 0.58 0.94 0.74 GPIR the best
Porosity (%) 51 53 45 to 75 GPIR within the limits
Service temperature (°C) 1220 1450 1150 to 1261
GPIR within the limits
Thermal conductivity @1100 °C ( W.m-1K-1) 0.462 1.068 2.6 – 2.8 GPIR the
best
93
The comparison shows that the geopolymeric porous insulating refractory (GPIR)
values comply with the specifications for porous insulating refractories.
The outcome of this study provided an insulating refractory material to be used in
applications up to 1250 °C. The geopolymeric methods drastically increased the
strength of the insulating refractory materials to make automatisation of the process
possible. The process is rapid and the in situ foaming of the geopolymer results in
high closed porosities and thus good thermal conductivities (Figure 5.1). In situ
forming of monolithic geopolymeric porous insulating refractory, benefits material
storage, handling and mould availability.
Figure 5.1: Thermal conductivity of insulating fire brick and insulating castables (Carniglia & Barna, 1992). STL indicating the Service Temperature limit (in °F) of the Insulating Fire Brick (IFB) and Geopolymeric Porous Insulating Refractory (GPIR)
The process is economical, as the major raw materials are inorganic waste
materials. The use of these materials further helps to resolve the problem of waste
disposal and benefits the environment.
94
The utilisation of these inorganic waste materials for the manufacturing of light-
weight porous insulating refractory materials, should lead to the preservation of
natural resources. The environment will benefit from this utilisation as the demand
on mining activities for raw materials will be reduced and discarded waste that
degrades the environment will be utilised.
A geopolymeric porous insulating refractory was successfully developed using the
inorganic waste material, coal fly-ash, as the main ingredient. Phosphogypsum and
iron-rich waste is not utilised in the manufacturing of insulating refractory materials
as it impacts negatively on the thermal properties of the insulating material. These
other waste materials are successfully incorporated in water filters from inorganic
waste materials – another project within the research group. This developed
insulating refractory material complies with the specifications (Table 5.1) laid down
for a porous insulating refractory material. The developed process delivers a very
good product without shifting the waste disposal problem by creating more or
different waste products detrimental to the environment and the developed
manufacturing process is economical.
The specific objectives of the project were all achieved, namely
• Characterising inorganic waste materials with regard to their physical and
chemical properties relevant to refractories.
• Investigation of different mixtures to result in a mix formulation complying with the
laid down specifications and
• An economical manufacturing route for porous insulating refractory materials
made of inorganic waste was found that lend itself to the automatisation of the
manufacturing process.
95
5.2 RECOMMENDATIONS By employing the following process steps in the geopolymerisation of the inorganic
waste materials a very good economical porous insulating refractory material would
be available on the market.
• Dissolve 10.53 % sodium hydroxide into 10.53 % of water. This reaction is
exothermic.
• Leave the solution to cool.
• Add 15.78 % sodium silicate.
• Add 10.53 % of meta-kaolin to the solution.
• Stir for 5 minutes.
• Add 55.63 % coal fly-ash while stirring the mixture.
• Add 0.005 % aluminum powder as the porogen.
• Cast the geopolymeric material into position or into moulds.
• Setting will occur within 3 hours.
• Leave to dry at room temperature for ± 24 hours.
• A heating procedure of 2.5 °C per minute may follow up to service temperature of
the product.
• The shrinkage which will occur will be as little as 0.3 % allowing for the formwork
to be removed if needed before firing.
The hypothesis of developing a geopolymeric porous insulating refractory using
inorganic waste materials as the main ingredient was successfully accomplished.
The successfully developed porous insulating refractory material from inorganic
waste materials is a suitable candidate to replace fireclay insulating materials or
other insulating materials in high temperature application vessels, used in industry.
The newly developed procedure is cost effective and manufacture can be easily