Case Study
16
o o
O c
0
O ,<Ł
December
2007
A Novel Way to Upgrade the Coarse Part of a
High Calcium Fly Ash for Reuse Into Cement Systems
Abstract
Reject fly ash (rFA) represents a significant portion of the fly ashes produced from coal-fired power plants.
Due to the high carbon content and large particle mean diameter, rFA is not utilized in the construction sector
(for example as supplementary cementing material) and is currently dumped into landfills, thus representing
an additional environmental burden. Recently, the feasibility of using rFA in a relatively small number of
applications, like solidification/stabilization of other wastes, has been investigated by different researchers.
However, as the overall amount of fly ash utilized in such applications is still limited, there is a need to
investigate other possibilities for rFA utilization starting from a deeper understanding of its properties. In the
work presented herein, mechanical and hydration properties of cementitious materials prepared by blending
the coarse fraction of a lignite high-calcium fly ash with ordinary cement were monitored and compared with
the respective ones of a good quality fly ash-cement mixture. The results of this work reveal that a relatively
cheap, bilateral classification-grinding method is able to promote the pozzolanic behaviour of the rEAs, so that
the overall performances of rFA containing cements are drastically improved. The evaluation of these results
supports the belief that appropriate utilization of non-standardized materials may lead to new environmental-
friendly products of superior quality.
1. Introduction
During the last decades fly ashes produced in power plants have been commonly used in the construction
sector as additives either to ordinary cement or to ready-mixed, high-performance and lately self-consolidating
concrete (Brameshuber and Uebachs, 2003; Oh et al, 2002; Xie et al, 2002). The utilization of such by-products,
in addition to supporting sustainable development principles, contributes directly to the protection of the
environment by reducing the amount of wastes to be landfilled thus also preventing impact related to leachate
generation and migration (Cheerarot and Jaturapitakkul, 2004; Sajwan et al, 2003). Reductions of the amount
of residues to be landfilled along with partial substitution of common raw materials used in cement making
are the major beneficial effects of the recycling process (Filipponi et al, 2003; Bijen, 1996). Normally, cement
industries utilize good-quality, or at least suitable-quality, fly ashes for manufacturing cement and concrete.
Fly ash suitability is determined from its compliance with specified criteria, such as unburned carbon, reactive
silica and alumina, and sulfur trioxide and FA fineness. Fly ashes can comply with the current European
standard EN 450-1 or with National Specifications, the latter ones published in order to also classify the fly
ashes that do not comply with the EU standard. The possibility however, that electricity producing companies
will adjust their productivity plans according to the quality of the derived by-product is unrealistic. In view of
the above considerations, each producing country has determined not only the quality but also the variability
range of the inorganic components of the local coal.
In general, fly ash utilization rate is mainly governed by: (i) the annual ash production levels in each country,
(ii) the FA demand of local cement and concrete industry and finally (iii) the qualitative variation of the
product with regard to the specifications in force. With the above criteria in mind, it is understandable that
Germany and the Netherlands present impressive fly ash utilization rates (that is almost equal to 100%) whilst
in other countries, including Greece, the corresponding rates remain low (approximately 10-12%). In Greece
efforts have been made during the last years to promote ash utilization. The most significant of those is a draft
proposal with specifications regarding the use of high-calcium fly ash (which is the local paramount solid by-
product) in non-reinforced concrete. Even though the clearest distinction between suitable and reject fly ashes
is derived from the conformity or not with specifications that are in force in each country, it seems that after
a size classification, ashes initially classified as reject could provide fractions that may meet the suitability
criteria. In any case, it is very important that the reuse of fly ash would increase every year. A practical way to
achieve this goal would be the application of a relatively inexpensive method that would enable the recovery,
upgrade and reuse of rFA.
Page 1 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
Coarse ashes, representing the most relevant subcategory of rFA, are composed of particles having a small
specific surface area and containing high levels of residual organic carbon. Apart from being less reactive,
such ashes are also characterised by problematic setting behaviour into the cement paste environment
(Ghosh and Sarkar, 1993). Despite being an arduous task, a respectable number of scientists have dealt with
low-quality ashes in an attempt to improve their unique assets. In previous investigations (Kiattikomol et
al, 2001), five different fly ashes were fractionated on a dimensional basis in order to investigate the effect
exerted by each different class on the chemical and physical properties of the cement mixtures. The results
revealed that the observed improvement in the strength development of concrete containing ground coarse
fly ash (GCFA) was mainly due to the pozzolanic behaviour exhibited by the GCFA itself; moreover, the
degree of the pozzolanic reaction increased with GCFA fineness (i.e., grinding degree). In another attempt
Jaturapitakkul et al (2004) managed to improve the efficiency of coarse ashes by inserting them into mortar
systems after reducing the average particle size to 3.8 (im. It was found out that concrete with about 25%
of GCFA attained excellent strength values, comparable to that attained by silica fume-concrete. A large
effort was also undertaken by CANMET on the mechanical and durability properties of high-volume fly ash
(HVFA) cement by using a coarse fly ash that did not meet the fineness requirements stated in ASTM C618
(Bouzoubaa et al, 2001). Their results showed that HVFA cement making technology offers an effective
way of utilizing coarse fly ashes that do not meet the criteria stated in the relevant American standard.
Other researchers studied the effect of specific additives on the properties of cement mixtures containing
the coarser fraction of high calcium fly ash, in order to highlight whether such a fraction could exhibit a
pozzolanic behavior. Yan and Yang (2000) for instance mixed low-quality fly ashes and waste fluorogypsum
to develop a new binder with satisfactory strength and high-volume stability. The mixtures were prepared
with fly ash-fluorogypsum-cement ratios (w/w) of approximately 45:40:15.
The main objective of this investigation is to ascertain the effectiveness of a new way of processing the
coarse part of a high-calcium fly ash. The authors were initially motivated by the fact that there is still no
clear picture of whether or not the coarser fraction of a fly ash could be engaged into pozzolanic reactions.
Recently Antiohos and Tsimas (2002) have proved that the coarser fractions of fly ashes contain much
more amorphous silica than the fine ones. Since amorphous silica is actually the main carrier of pozzolanic
reactions, it is believed that if the coarse part of the fly ash could be first segregated (with the aid of an
air classification process) and then ground in order to take advantage of the "filler" effect, this could lead
to the production of a new type of ash of improved pozzolanicity and presumably better performance in
blended systems. If the above scheme proves to be efficient, it will represent a promising option for the use
of commonly discarded materials, providing a valid alternative to landfilling.
2. Experimental Program
2.1 Materials
The materials used in this investigation were a normal setting cement type I 42.5 and a high-calcium fly
ash from Ptolemais area (designated as 7^). The chemical composition and main physical properties of the
initial materials are given in Table 1. For estimating the oxide composition of the fly ash, the procedures
described in the European Standard EN 450-1 (Fly ash for concrete - Part 1) were used. The latter were also
used for determining the insoluble residue of the fly ash. The "Active ratio" shown in Table 1 is considered
as the ratio of active silica to the total silica in the fly ash. For calculating the content of the LOI-free fly
ash constituents soluble in hydrochloric acid and potassium hydroxide (Glass content of fly ash, coded
as S in Table 1) the procedure described in the RILEM recommendations (TC FAB-67 Use of Fly Ash in
Building) was adopted. A procedure also described in the abovementioned recommendations was used for
the determination of the free CaO content of the fly ash. Briefly, one gram of the original ash was mixed
with ethyl acetoacetate and isobutanol and then heated for 1 hour. The mixture was then filtered through a
filter crucible and the filtrate was titrated with 0.1 N hydrochloric acid. Finally, the specific gravity of the
fly ash was determined using a standard Le Chatelier flask according to ASTM C-188.
Page 2 of 12
-------
Case Study
16
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
2.2 Description of the refinement process
Bulk fly ash was fractionated using a lab classifier (alpine apparatus) with a sieve measuring 45 (im followed
by a cyclone, the latter aimed at collecting the underflow stream.
Table 1 Chemical composition (% by mass) and main physical characteristics of raw materials
Cement 7F 7FP 7FM
CaO
CaOf
CaCve8
SiO2
SiO2rea
AI203
Fe203
MgO
S03
Na2O
K2O
LOI
Ysd
IR (%)*
Glass content. Sb(%)
Elaine Fineness (cm2/g)
Specific Gravity
65.01
0.63
n.ac
20.28
n.a
4.75
3.76
1.61
2.55
0.17
0.35
2.31
n.a
0.18
n.a.
3.760
3.13
29.79
4.05
25.13
36.92
29.13
13.50
7.06
2.69
5.10
0.92
0.50
4.36
78.90
14.52
85.48
5.450
2.80
21.02
2.36
16.35
42.79
32.46
14.92
5.82
3.32
3.09
0.74
0.56
4.66
75.86
15.69
84.31
5.420
2.92
33.02
4.45
27.43
33.50
26.49
11.80
5.70
3.43
8.79
0.71
0.61
4.01
79.07
10.40
89.60
5.250
2.78
The method specified in the European Standard EN-450 was followed for the estimation of the
reactive silica and calcium oxide contents and the insoluble residue (IR) of the fly ashes.
b The method specified in the RILEM Recommendations (TC FAB-67 Use of Fly Ash in Building) was
followed for calculating the content of the LOI-free fly ash constituents soluble in hydrochloric acid and
potassium hydroxide (S =100-IR).
0 n.a: not available
d Ys: active ratio (ratio of active: total silica)
Using the above refinement process, two new fly ashes were produced: a coarse one (approximately 58%
by weight of the bulk ash remained on the sieve, designated as T^) and a fine one (with particle size less
than 45 (im, coded here as ^FM). Chemical composition and physical characteristics of the fractionated
ashes were estimated with the use of the same procedures that were followed in the case of the bulk ash
(described in detail in 2.1.) and are also provided in Table 1. To neutralize the effect of fineness on their
reactivity, all fly ashes (i.e., the bulk and the two fractions) were ground (different retention times were
applied in each case) in a lab ball mill to reach similar particle size distributions, as shown by Elaine
fineness measurement (according to ASTM C-204) and by the data shown in Fig. 1 (determined by means
of laser granulometry).
Page 3 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
100
ra
a.
|
O
50
*T5»
-TV, —
1,5 2
8 12 16 24 32 48 64 96 128 192
Sieve I
Figure 1 Particle size distribution ofTf Tm and Tfp samples
2. 3 Preparation of
specimens and testing
In order to have a first indication
of the pozzolanic behavior
of rp TFM and Tfp fly ash, the
Fratini test was carried out on
different suspensions containing
the cement-fly ash mixtures
(fly ash to cement ratios equal
to 1:5 and to 1:2.5). The test
involves accelerated curing of
the suspensions for 8 days at
40°C and RH=100%. At the
end of the curing period, the
suspension is filtered through a
filter crucible and analysed for
Ca content (titration performed
with HC1 0.1 N) and alkalinity
(titration with standard EDTA
solution). Comparisons are made
using the solubility curve of Ca(OH)2. Each mixture was tested twice and the mean values are reported.
Compressive strength development was monitored on mortar specimens prepared following a procedure
described in earlier works (cementitious materials-to-sand ratio of 1:3 and W/CM of 0.5) (Antiohos et al,
2004; Papadakis et al, 2002). In the present study 20% and 40% replacement of cement by fly ash were
adopted and compression tests were conducted at 2, 7, 28, 90 and 365 days after mixing. A mortar with no
fly ash was also prepared (control) for comparison. Strength results were used to calculate the efficiency
factors (^-values) in each case. To simulate paste into mortar, paste specimens were also prepared and
cast in plastic vials after being intensively shaken to remove any air content. At the same time intervals
adopted for the compressive strength measurements (with the exception of 1 year old samples), the samples
were broken into pieces, treated with organic solvents (acetone and diethylether) so as to stop any further
hydration, and dried by means of a vacuum pump. The dried fragments were further pulverized to less
than 125 (im and kept sealed until testing. Testing involved thermogravimetric analysis for monitoring the
evolution of the pozzolanic reaction. TG measurements were performed into a platinum crucible (volume
equal to 70 (il) by means of a Mettler STARe 85 l/LF/1600 TG operated in nitrogen atmosphere (50 ml min~
!) at a heating rate of 10°C min'1 from ambient temperature to 1000°C. The phase composition of powder
XRD characterization of the fractionated ashes and the detection of possible alterations in the nature of
hydration products was carried out by means of a Siemens D 5000 X-ray diffractometer (CuKa radiation,
40KV, 30mA) operated at 0.02°/sec in a 26 scale. Identification of the diffractogram was carried out using
a Diffrac-AT Database (PDF-release year 2000).
3. Results and Discussion
3. 1 Properties of fractionated ash samples
Although the effect of classification processes on the chemical composition of fly ash has been questioned
by numerous researchers (Erdoglu and Turker, 1998; Berry et al, 1989) a close look at the data of Table
1 reveals that the individual fractions differ substantially with regards to the percentage of some of their
principal constituents. This is of primary importance since differences in the chemical composition are
expected to influence the pozzolanic potential of each fraction. The coarser fraction (7^) for instance is
notably enriched in total and active silica, but its free CaO and SO3 contents are closer to the requirements
(not greater than 2.5% and 3% respectively) stated in the European standard EN 450-1, if compared to
Page 4 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
Figure 2 XRD patterns of fractionated Tfp and Tfufly ash samples (Q: Quartz, L: Lime, Cr:
Cristobalite, A: Anhydrite, M: Mullite, MG: Magnetite, W: Wollastonite, V: Vaterite, CA:
C3A, An: Anorthite, Gl: Gehlenite, G: Gismondine, P: Portlandite)
the bulk ash. On the contrary, the
TfM fraction contains much more
CaO (as this is concentrated in the
smaller size particles), but also
unusually high percentages of
free CaO and SO3. With respect
to the last two parameters, the
use of ground coarse fly ash may
be advantageous since both free
CaO and SO3 may be harmful to
mortar and concrete durability.
The loss-on-ignition (LOI) values
of the fractioned samples did not
change appreciably compared to
that of the initial fly ash. In other
works (Berry et al; 1989), the finer
ash (obtained by classification)
is reported to show higher LOI
values than the original bulk fly
ash. This is very encouraging since higher LOI is usually (but not always correctly) associated with high
unburned carbon that causes undesired swelling effects in the paste. No significant differences were detected
regarding the glass phase content of T^ and its fractions, while a notable increase in the specific gravity
value of Tfp was attributed to prolonged grinding that caused the breaking of cenospheres and plerospheres
(Mora et al, 2003). Finally, from the mineralogical point of view (Fig. 2) it can be qualitatively observed
that in the Tfp fraction more silica compounds like quartz, mullite and cristoballite were detected and lime
was hardly traced, whilst intense reflections for the presence of CaO were detected in the respective finer
fraction. The pattern of TFM diffractogram reveals higher contents of anhydrite and C3A compounds, in
accordance with the abundant presence of CaO.
3.2 Pozzolanicity test results
Results of the Fratini test are presented in Fig. 3 both for the 20% and 40% fly ash addition. It is known
that a measured point below the solubility curve of Ca(OH)2 is an indication that the examined suspension
is undersaturated in Ca(OH)2
due to the
produced
fact that Ca(OH)2
during cement
hydration has been consumed by
pozzolanic reactions. Obviously,
when a measured point is
above the saturation curve, it is
concluded that the generation of
Ca(OH)2 from cement hydration
is greater than the equivalent
amount of Ca(OH)2 that reacts
with the material inside the
suspension. If so, the examined
material can hardly be described
as pozzolanic. From the data
shown in Fig. 3 it can be seen
that almost all mixtures tested
presented pozzolanic behavior
especially at high levels of
cement substitution (i.e., 40%
18
16
14
-7 12
10
6]
2
10 20 30 40 50
Total alkalinity (mmol OHVL)
60
70
Figure 3 Fratini test results for fly ash — cement mixtures (FA content equal to 20% and 40%
respectively)
Page 5 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
FA content). The only measured point that was found to be over the saturation curve corresponds to the
mixture containing 20% of the fine rFM ash. This is probably due to the very low active silica content of
this sample. On the other hand both mixtures incorporating the ground coarse fraction 7^, exhibit superior
pozzolanic action even when compared to the cement prepared with the bulk ash. This is the result of its
very high active silica content which is the main binder of available Ca(OH)2 and the factor that governs
the pozzolanic performance of the ashes tested. Despite the encouraging results presented here, they should
only be used as a first indication for the future performance of the corresponding pastes and mortars since
the Fratini test was performed under controlled-accelerated conditions and it cannot predict the actual
behavior of the same samples under normal hydration conditions.
3.3. Mechanical Properties
Compressive strength development of the control and fly ash mortars is presented in Table 2. During the first
week of hydration, specimens containing the bulk T^ develop strength slower than the control since at this
stage ash acts mostly as an inert material. The rate of strength development increased significantly for the
two samples containing the T^ fraction, which outperformed the reference specimen as early as two days
after mixing. In contrast, TFM inclusion retarded notably the strength gain of mortars. Considering that the
particle size distribution of the two processed samples is practically the same, the best performance observed
for the samples containing the 7^, was ascribed to the different chemical composition, and specifically to
the higher contents of active silica and alumina which are the main carriers of the pozzolanic reactions.
Table 2 Compressive strength development of reference and fly ash mortar samples
Compressive strength (MPa)
Control
20 TF
20TFM
20TFP
40 TF
40TFM
40TFP
2
24.7
19.4
14.0
25.8
12.8
10.6
22.1
7
39.2
38.0
28.0
38.3
26.8
16.9
33.3
Age (days)
28
50.6
49.4
46.6
51.7
42.8
39.2
49.0
90
59.7
59.7
58.0
62.7
57.3
55.6
60.3
365
68.0
68.4
67.2
71.6
64.8
68.0
70.3
As hydration evolves, 7^, mortars continue to show better performance than the other ash-containing samples,
reaching emphatic values of 62.7 and 71.6 MPa after three and twelve months of curing respectively. The
difference observed in the strength development grows larger with curing time, and becomes high after
the first month, since during this stage the role of active silica is known to be predominant (Antiohos et
al;2004).
When cement replacement is increased to 40%, the strength values were normally diminished. This decrease
is certainly more pronounced in the case where the bulk ash and its fine fraction were utilised, indicating
that those ashes (especially the latter) may not be used for applications where immediate need for strength
development is required. It becomes evident that reactive constituents contained in those ashes cannot
compensate, at least during the initial period of hardening, for such an absence of clinker's strength carriers.
Despite the significant cement replacement, the ground coarse T^ ash performs again impressively from the
first week of curing. With the progress of hydration, Tfp mortars develop strength faster than all the other
Page 6 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
specimens tested (including the control one) due to the full commencement of the pozzolanic reactions. In
fact, after one month of hydration, T^ mortar exhibits similar strength to the non-fly ash specimen, while at
the same stage, it outperforms the sample containing the original ash. The strength difference between the
ground coarse ash-containing samples and the other samples is continuously growing towards the end of the
curing period (1 year), thus confirming once again the observations made by the authors on the significance
of active silica for long-term strength evolution.
3.4 Efficiency factors
It has been well established (Papadakis and Tsimas; 2001) that in the case of mortars and concrete that
incorporate supplementary cementing materials, the k-value is derived from the following expression for
the measured compressive strength (/"):
wi(c+kP)
-a)
where K is a parameter depending on the cement type (here 38.8MPa), C and P are the cement and fly ash
contents respectively in the mortar (kg/m3), W is the water content (kg/m3) kept constant in all the mixes
and a, a parameter depending mainly on curing conditions (thus its numerical value for the tested samples
changes with curing time). Using the mean measured values of the compressive strength of the control
specimen, the parameter a was estimated as 1.36, 0.98, 0.69, 0.46 and 0.24 for 2, 7, 28, 90 and 365 days,
respectively. Based on the above expression and strength values shown in Table 2, the ^-values of the bulk
and fractionated ash blends were calculated and presented in Table 3. Data in Table 3 reveal once again
the inability of the bulk and fine ashes to act drastically from the start of the curing process and the high
reactivity of the ground coarse fraction. The latter exhibits higher k-values than the control mortar (in the
case of moderate cement replacement) and in the case of severe cement replacement (i.e., 40%) it reaches
unity very quickly and it exceeds it at the end of the curing process.
Table 3 Efficiency factors of fly ash samples
20 TF
20TFM
20TFP
20 Tpp (theor-v)
40 TF
40TFM
40TFP
2
0.66
0.31
1.07
0.62
0.55
0.92
7
0.92
0.28
0.94
0.60
0.28
0.81
/(-values
Age (days)
28
0.92
0.74
1.07
1.05
0.75
0.63
0.95
90
1.00
0.89
1.19
1.23
0.92
0.87
1.02
365
1.01
0.93
1.21
1.41
0.89
0.99
1.03
a Theoretical values derived using equation (2)
In a recent work Papadakis et al. (2002) have reported, for the first time, analytical expressions that related
active silica of artificial pozzolans with k-values of their respective cementitious systems aiming to enable
a first approximation of their performance starting from the amount of amorphous silica. The authors
concluded that for a cementitious system containing SCMs (supplementary cementing materials), ^-values
can be expressed as follows:
Page 7 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
where: Fs is the weight fraction of the oxide SiO2 in the SCM (given in Table 1), which contributes to the
pozzolanic reactions (i.e. the ratio of active silica to the total silica in the SCM) and fS,P and fS,C are the
weight fractions of silica in the SCM and cement respectively. By applying the above equation in the case
of 207^ blended cement after the first 28 days (a stage where active silica is known to hold a critical role),
almost identical ^-values (with the experimental ones) were calculated. The latter observation denotes the
consistency of the results from the theoretical model even when reject fly ashes are considered.
3.5 Pozzolanic reaction evolution
By estimating the available Ca(OH)2 that exists in a cementitious matrix at different stages of curing, the rate
at which pozzolanic reactions are taking place can be obtained. For the ashes used herein, Table 1 shows that
their free lime contents are practically the same. Therefore the amount of secondary portlandite generated
by fly ash hydration would be the same for all ashes tested. Based on this assessment, thermogravimetry
results were used to evaluate the progress of pozzolanic reactions for the cements that performed best from
the mechanical point of view, that is, for a 20% ash addition. The results were then inserted in the following
equation proposed by Paya et al. (2003) to calculate the percentage of fixed lime (FL), a factor directly
associated with the progress rate of the pozzolanic reaction (results are demonstrated in Fig 4):
Fixed Lime (%) =
(CHc-C%)-CHPxl()0
CH. C«
where CHc is the CH content of the control paste for a given curing time, CHp is the CH content of the fly
ash-cement (FC) paste at the same age and C is the proportion of cement in the examined paste.
In all cases fixed lime values increase with curing time. Negative values observed for the fractionated ash
pastes after 2 days of hydration are indicative of the excess of CH created from the instant hydration of their
free lime and of the incapability
of both ashes to contribute from
the beginning of hardening.
The inclusion of the coarse
rF ash in the cement system
brings forward positive results,
especially after the first week,
as manifested by the impressive
FL value measured at 28 days
(more than 18% compared to
around 6% measured at the same
age for the other ash samples).
Throughout the examined curing
period (up to 90 days), the cement
containing the ground coarse ash
fraction exhibits higher fixed
lime percentages than the other
blended cements as a result of
the higher active silica contained
in the specific fraction. This is
increasingly liberated from the
core of the ash to participate in
the reactions with available lime
to form additional, secondary
C-S-H gel that brings about a
Figure 4 Fixed lime values for Tf Tfp and TFU—containing mixtures (FA content equal to 20%)
Page 8 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
noticeable strengthening of the paste (the corresponding mortar exhibited a very high k-value of 1.17).
3.6 Hydration products identification
XRD measurements confirmed the observations of the thermogravimetry analysis regarding the nature of
newly formed products during hydration. Significant presence of portlandite and ettringite was confirmed in
the XRD patterns of the fly ash pastes after 28 days of curing (Fig. 5). It can be easily seen that in every paste
tested, crystalline Ca(OH)2 was generated along with ettringite as a result of the strong presence of sulfur
trioxide (mainly in the form of anhydrite) in the ashes used. In the recent past, Tishmack et al. (1999) have
noted that ettringite formation increases with increasing anhydrite contents. This is probably the case here
regarding the pronounced presence of AFt phases in all pozzolanic specimens, especially those containing the
finer fraction (7^) which contains much more anhydrite than the other samples (Table 1 and Fig. 2). The latter
sample is the only one where traces of monosulfate hydrate were detected, possibly due to the dissolution of
SO3 ions from ettringite. On the samples with the coarser fraction of T^ the production of hydrated calcium
aluminate hydrate phases can be
clearly detected from the second
week of curing, as well as some
gismondine. Phases like these
are both contributing towards
filling the pores in the matrix and
subsequently strengthening the
paste. This is in accordance with
compressive strength results
shown in previous sections.
Inherent crystalline constituents
of fly ashes, such as lime and
quartz appear also in the patterns
of their hydrated blended pastes,
whilst presence of calcite in all
systems could be attributed also
to possible carbonation during
handling. A decrease in the
intensity of the C2S and C3S peaks
with age corresponds to the
acceleration of their hydration
in the presence of finely ground
fly ash, as clearly explained by
Berry et al. (1989). The presence
of ettringite is increased with the
ash addition (i.e. 40%, Fig. 6),
contrary to portlandite whose
reflections are less intense as a
result of the higher CH depletion
after one month (coinciding with
the high FL values presented
earlier). In this case where
an extra 20% of fly ash was
added, the presence of calcium-
aluminate (CA) and gismondine
phases in the ground coarse fly
ash sample is more pronounced,
whereas such compounds are not
Mk^uJ*U^
E' "' Et I ' p ^FM-28d
^
iuJ^^/^^
i^^^
Figure 6XRD patterns ofTfu Tf and TFp —containing cement pastes after 28 days of
hydration (FA content equal to 40%) (Et: Ettringite, CA: Calcium aluminium oxide hydrate,
G: Gismondine, P: Portlandite, Cc: Calcite, Q: Quartz, G: Gypsum, A: Alite, Be: Belite, AN:
Anhydrite, AFM: Calcium aluminum sulfate hydrate)
Page 9 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
detected at all in the respective samples with the fine fraction ash. It is possible that the higher percentages
of soluble silica and alumina of T^ are responsible for the formation of such phases that contribute to the
hardening of the paste.
4. Conclusions
In this paper, a novel way of utilizing reject coarse fly ashes was introduced. A bilateral method comprised
of air-classification and further grinding of the coarse fraction of a lignite high-calcium fly ash was applied
and evaluated. This effort was primarily motivated by previous experiments that revealed the enrichment of
the coarse fractions of fly ashes in total and soluble (thus reactive) silica and a related increase in pozzolanic
reactivity. In general the proposed method could be applied in the case of: (i) reject fly ashes aiming at
recovering fractions that comply with the specifications in force, but also of (ii) good quality fly ashes
that fulfill the conformity criteria, focusing at reusing specific fractions of advanced quality for certain
applications.
It is noted that the authors have also worked in the same direction, that is, the upgrade and reuse of the
reject part of fly ash, with fly ashes of low-lime content. The preliminary results from this attempt are very
convincing also with regards to the potential use of initially rejected fractions of this valuable by-product.
Moreover, given the possible retarding or accelerating (depending on their composition, chemical form and
typology of the associated anion) effect of the heavy metals that may be contained in fly ash, the authors
are currently investigating the presence of trace metals in the separated fractions of this work in order
to clarify whether and to what extent their presence could be partly responsible for the behavior of each
different fraction examined. With regards to the main findings of the present work, these are summarized
as follows:
1. The air-classification process verified previous findings on the enrichment of the coarser fly ash
fractions in total and active silica and of the finer ones in CaO, free lime and sulfur content. The
corresponding decrease of the last two parameters, the relatively small increase of LOI and the higher
pozzolanic potential (as demonstrated with the aid of the Fratini test) of the coarser fraction, might
enable the incorporation of ground coarse fly ashes (GCFA) into cement and concrete. It is concluded
that by applying the described technique a significant percentage (in this work approximately 58%)
of fly ash could be upgraded and recycled into systems of advanced performance.
2. Inclusion of GCFA into cement mortars increased notably the rate of strength development from the
beginning of the curing period, mitigating one of the biggest drawbacks associated with the utilization
of fly ash. Even when large volumes of cement were substituted, rFp-based systems exhibited strength
superiority indicating that they may be used in the production of High-Volume FA Concrete or even
High-Strength Concrete. On the other hand, the sole use of the fine fraction of the bulk ash could
not present sufficient early strength; therefore its use should be excluded from structures that require
adequate strength development. However, it can still be reused as an aggregate for lightweight and
normal concrete or even as a road base material and flowable fill.
3. Given that the particle size distribution and specific surface of the two processed samples is practically
the same, it is believed that the superiority of the coarser fraction is attributed to the differences in
the chemical and mineralogical composition of the samples, mostly the excess of active silica and
alumina which are the main carriers of the pozzolanic reactions.
4. Strength superiority of ground coarse fraction was testified also by means of lvalue. A previously
reported expression, correlating active silica of artificial pozzolans with ^-values, was also validated
in the present work. Using such an expression can lead to a relatively safe approximation of the future
mechanical performance of the final product.
5. High FL values were measured for the coarser fraction based mortars. FL values remained notably
higher than those of the other blended cements throughout the examined period. Soluble silica is
Page 10 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
increasingly liberated to participate in the reactions with available lime to form additional, secondary
C-S-H gel, causing the strengthening of the paste.
6. C-S-H gel, portlandite and calcium sulphoaluminates (mainly ettringite) were the main hydration
products formed in the GCFApastes. Peaks of gismondine were also detected after the first two weeks
of hydration. Generation of these products is favored by the increasing liberation of amorphous silica
that abounds in the coarse part of fly ash. Such compounds are considered partly responsible for the
compaction of the paste and subsequent strength improvement.
5. References
Antiohos, S., Papageorgiou, A., Tsimas, S., 2004. Activation of fly ash cementitious systems in the presence of quicklime. Part II - Nature of
hydration products, porosity and microstructure development, Cement and Concrete Research (submitted).
Antiohos, S. and Tsimas, S., 2002. Reactive silica of fly ashes in relation to the burning condition of lignites, in: Dhir, R.K. (ed.), Proceedings of the
International Conference Challenges of Concrete Construction, Dundee, Scotland, 2002, 71-80.
Antiohos, S. and Tsimas, S., 2004. Investigating the role of active silica on the hydration mechanisms of high-calcium fly ash/cement systems,
Cement and Concrete Composites 27, 171-181.
Berry, E.E., Hemmings, R.T., Langley, W.S., Carette, G.G., 1989. Beneficiated fly ash: hydration, microstructure and strength development in
Portland cement, In: Proceedings of the 1st International Conference on Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete, SP-114,
Detroit, USA, 241-273.
Bijen, J., 1996. Benefits of slag and fly ash, Construction and Building Materials 10, 309-314.
Bouzoubaa, N., Zhang, M.H., Malhotra, V.M., 2001. Mechanical properties and durability of concrete made with high-volume fly ash blended
cements using a coarse fly ash, Cement and Concrete Research 31, 1393-1402.
Brameshuber, W. and Uebachs, S., 2003. The Influence of Air Voids on the Properties of Self-Compacting Concretes Containing Fly Ash. In:
Wallevik, O. and Nielsson, I. (Eds.), Proceedings of the 3rd International RILEM Symposium on Self-Compacting Concrete, Reykjavik, Iceland,
870-880.
Cheerarot, R. and Jaturapitakkul, C., 2004. A study of disposed fly ash from landfill to replace Portland cement, Waste Management 24, 701-709.
Erdoglu, K. and Turker, P., 1998. Effects of fly ash particle size on strength of Portland cement fly ash mortars, Cement and Concrete Research 28,
1217-1222.
Filipponi, P., Polettini, A. Pomi, R. and Sirini, P., 2003. Physical and mechanical properties of cement-based products containing incineration
bottom ash, Waste Management 23, 145-156.
Ghosh, S.N. and Sarkar, L.S., 1993. Mineral Admixtures in Cement and Concrete, in: Ghosh, S.N. (ed.), Progress in Cement and Concrete, ABI
Books, New Delhi, India, 118-157.
Jaturapitakkul, C., Kiattikomol, K., Sata, V. and Leekeeratikul, T, 2004. Use of ground coarse fly ash as a replacement of condensed silica fume in
producing high-strength concrete, Cement and Concrete Research 34, 549-555.
Kiattikomol, K., Jaturapitakkol, C., Songpiriyakij, S. and Chutubtim, S., 2001. A study of ground coarse fly ashes with different finenesses from
various sources as pozzolanic materials, Cement and Concrete Composites 23, 335-343.
Mora, E., Paya, J., Monzo, J., 2003. Influence of different sized fractions of a fly ash on workability of mortars, Cement and Concrete Research
23, 917-924.
Oh, B.H., Cha, S.W., Jang, B.S. and Jang, S.Y., 2002. Development of high-performance concrete having high resistance to chloride penetration.
Nuclear Engineering and Design 212, 221-231.
Papadakis, V.G., Antiohos, S., Tsimas, S., 2002. Supplementary cementing materials in concrete - Part II: Afundamental estimation of the efficiency
factor, Cement and Concrete Research 32, 1533-1538.
Papadakis, V.G. and Tsimas, S., 2001. Supplementary Cementing Materials for Sustainable Building-Sector Growth, European Commission DGXII,
Marie Curie Fellowship, Final Report, Project No HPMF-CT-1999-00370, NTU of Athens, Greece, p. 74.
Paya, J., Monzo, J., Borrachero, M.V., Velasquez, S., 2003. Evaluation of the pozzolanic activity of fluid catalytic cracking catalyst residue (FC3R).
Thermogravimetric analysis studies on FC3R-Portland cement pastes, Cement and Concrete Research 33, 603-609.
Sajwan, K.S., Paramasivam, S., Alva A.K., Adriano, D.C. and Hooda, P. S., 2003. Assessing the feasibility of land application of fly ash sewage
sludge and their mixtures, Advances in Environmental Research 8, 77-91.
Tishmack J.K., Olek J. and Diamond S., 1999. Characterization of high-calcium fly ashes and their potential influence on ettringite formation in
cementitious systems, Cement and Concrete Aggregates 21, 82-92.
Xie, Y., Liu, B., Yin, J. and Zhou, S., 2002. Optimum mix parameters of high-strength self-compacting concrete with ultrapulverized fly ash.
Cement and Concrete Research 32, 477-480.
Yan, P. and Yang, W., 2000. The cementitious binder derived with fluorogypsum and low-quality of fly ash, Cement and Concrete Research 30,
275-280.
Page 11 of 12
-------
A Novel Way to Upgrade the Coarse Part of a High Calcium Fly Ash for Reuse Into Cement Systems
Case Study
16
Submitted By:
S. K. Antiohos, S. Tsimas
National Technical University of Athens, School of Chemical Engineering
9 Heroon Polytechniou, Zografou Campus, GR-157 73 Athens, GREECE
Tel: +30 210 772 3095, Fax: +30 210 772 1727, e-mail: stangits@central.ntua.gr
First-author:
Stylianos K Antiohos
National Technical University of Athens, School of Chemical Engineering
9 Heroon Polytechniou, Zografou Campus, GR-157 73 Athens, GREECE
Tel: +30 210 772 2893, Fax: +30 210 772 3188, e-mail: adiochic@central.ntua.gr
Technical University of Athens, School of Chemical Engineering
9 Heroon Polytechniou, Zografou Campus, GR-157 73 Athens, GREECE
COAL COMBUSTION
PRODUCTS PARTNERSHIP
This coal ash utilization case study is a selection of the Coal Combustion Product Partnership. For
more information, consult C2P2 web site at http://www.epa.gov/epaoswer/osw/conserve/c2p2/
Page 12 of 12
------- |