Construction and Building Materials 183 (2018) 226–233

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Controlling workability in alkali-activated Class C fly ash

Watanyoo Rakngan a, Trevor Williamson a, Raissa D. Ferron a, Gaurav Sant b,
Maria C.G. Juenger a,⇑
a
Department of Civil, Architectural, and Environmental Engineering, The University of Texas at Austin, 301 E. Dean Keeton St. C1748, Austin, TX 78712, USA
b
Department of Civil and Environmental Engineering, University of California, Los Angeles, CA 90095, USA

h i g h l i g h t s

 Alkali-activated Class C fly ashes have high strength but poor workability.
 Retarders, silica modulus, and mixing intensity can change workability.
 Workability modification can reduce strength, particularly when retarders are used.

a r t i c l e i n f o a b s t r a c t

Article history: Alkali-activation of Class C fly ashes can produce Portland cement-free binders that take advantage of
Received 18 April 2017 sodium aluminosilicate hydrate (N-A-S-H) and aluminum-substituted calcium silicate hydrate
Received in revised form 18 June 2018 (C-A-S-H) phases for strength development. However, alkali-activated Class C fly ashes suffer from poor
Accepted 20 June 2018
workability and rapid setting; traditional water-reducing admixtures are not effective. In this study,
different methods to modify the workability of alkali-activated Class C fly ash mixtures were investi-
gated. Retarders decreased the mixture stiffness and delayed setting, but strength development suffered.
Keywords:
Increasing mixing intensity slightly improved workability without significantly impacting strength.
Geopolymer
Alkali-activated materials
Controlling silica modulus significantly improved workability, but the impact on strength was
Fly ash inconsistent.
Workability Ó 2018 Elsevier Ltd. All rights reserved.
Slump
Retarder
Shear mixing

1. Introduction Most research studies on alkali-activated fly ash are based on

Alkali-activated materials (AAM) are a class of binders for con-
crete that contain no Portland cement, but instead consist of alu-
minosilicate powders mixed with alkaline solutions. They can be
categorized based on the composition of the system. When the
source materials contain primarily silicon and aluminum (e.g.
metakaolin, some Class F fly ash, and some natural pozzolans),
the primary reaction product in the hardened material is a sodium
aluminosilicate hydrate phase (N-A-S-H). N-A-S-H phases are
sometimes called ‘‘geopolymers,” a term that is sometimes used
to refer to the AAM as a whole. When the source materials contain
calcium along with silicon and aluminum (e.g. blast furnace slag,
some Class F fly ash, and Class C fly ash), the reaction products
contain a mixture of phases, including N-A-S-H phases,
aluminum-substituted calcium silicate hydrates (C-A-S-H), and
some crystalline phases [1].
⇑ Corresponding author.
E-mail address: mjuenger@mail.utexas.edu (M.C.G. Juenger).
low calcium, Class F fly ash [1]. Class C fly ashes, which generally
contain more than 20% CaO, have been shown to produce a stron-
ger concrete than Class F fly ashes when alkali activated [2] and a
denser microstructure with smaller pores [1]. However, alkali-
activated Class C fly ash often exhibits rapid stiffening and setting
[2,3], which can inhibit workability. Further, several studies have
demonstrated variability in the strength development of alkali-
activated Class C fly ash systems with changing fly ash and activat-
ing solution compositions [4–6].
Alkali-activated Class C fly ashes can be promising binders for
concrete if their workability problems can be overcome. In Port-
land cement systems, chemical admixtures (e.g. water reducers)
are frequently used to combat problems with workability. How-
ever, using water reducers and superplasticizers for AAM is com-
plicated because: (i) most are not stable in highly alkaline
solutions (pH > 13) [7], (ii) they have been shown in some cases
to have no direct effect on material consistency (e.g. mortar flow)
[8,9], (iii) they can negatively impact strength development [10],
and/or (iv) they have variable results depending on the activator

https://doi.org/10.1016/j.conbuildmat.2018.06.174
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 W. Rakngan et al. / Construction and Building Materials 183 (2018) 226–233 227

solution composition [11,12]. This necessitates further investiga- Table 1

tion of chemical admixtures or other methods of workability mod- Composition and Properties of Class C fly ashes.

ification for AAM. Oxide or P B R

Retarding admixtures can also be used in concrete to modify Property
workability. Collins and Sanjayan [13] examined the effects of an Al2O3 (%) 17.44 17.51 19.59
admixture containing sodium gluconate on alkali-activated blast SiO2 (%) 32.55 35.79 37.00
furnace slags. Workability was improved, but one-day compressive CaO (%) 28.95 26.89 23.88
Fe2O3 (%) 5.48 5.99 6.32
strength was reduced when this known retarder was used, partic- K2O (%) 0.46 0.56 0.54
ularly as dosage increased. Rattanasak et al. [14] examined the MgO (%) 6.45 6.25 4.76
effects of selected chemical admixtures on the setting time and Na2O (%) 1.86 1.96 1.73
compressive strength of alkali-activated high-calcium fly ash. The SO3 (%) 2.98 1.99 2.01
TiO2 (%) 1.32 1.15 1.42
changes in the initial and final setting times were negligible when
P2O5 (%) 0.89 0.81 1.20
CaSO4 was added to the pastes. Addition of Na2SO4 at dosages of d50 (lm) 11.68 11.34 13.48
1 wt% and 2 wt% considerably delayed the initial setting time,
but the final setting time was unaffected. Addition of sucrose at
ratios of 1.87–1.89. Similarly, the fly ashes had similar particle size
dosages of 1 wt% and 2 wt% did not affect the initial setting time,
distributions with median sizes of 11–13 lm.
whereas the final setting time was significantly delayed. Addition
A 4 M sodium hydroxide solution was used as the activating
of 1 wt% addition of CaCl2, CaSO4, Na2SO4 and sucrose increased
solution in this study. This concentration was chosen because it
the compressive strength of the mortars by 9–25%. However,
has been shown to provide adequate strength in geopolymer mor-
2 wt% addition of NaSO4 resulted in smaller increases in strength
tars [24]. It should be noted that higher concentrations can result
(3–6%) or even a strength reduction compared to the 1 wt% addi-
in higher strengths [24,25]. The solution was prepared by diluting
tion. Antoni et al. [15] examined the use of borax as retarding
reagent-grade 50 wt% NaOH solution (ACROS Organics) with ultra-
admixture for alkali-activated high calcium fly ash. Borax additions
pure water (Nanopure). NaOH solutions were allowed to equili-
of 1–7 wt% of fly ash extended the setting time of pastes compared
brate at room temperature for at least 24 h before use. When
to the control mixture without borax. The compressive strength
employed, most admixtures were dissolved in the activating solu-
also slightly increased with the addition of 1 wt%–5 wt% borax,
tion immediately prior to mixing of the pastes. Reagent-grade
while the compressive strength slightly reduced with the addition
chemicals used as admixtures included: sodium gluconate
of 7 wt% borax.
(Sigma-Aldrich), sodium tetraborate decahydrate (borax) (Sigma-
Addition of silica to the activating solution has also been shown
Aldrich), and anhydrous sodium sulfate (Fisher). Two commercial
to improve the workability of alkali-activated binders. For exam-
admixtures were also tested: Sikament N (Sika) and Recover
ple, increasing the sodium silicate to alkaline solution ratio from
(Grace). Sikament N is a high range water reducing admixture that
0 to 0.4 improved the flow and the compressive strength of
contains sodium naphthalene sulfonate as the main ingredient. It
alkali-activated slag pastes [16].
meets the requirements of ASTM C494 [26] as a Type A and F
In the work presented here, a variety of chemical admixtures
admixture. For general concrete applications, dosage rates of
and mixing methods were used to quantify the effects of these
390–1000 mL/100 kg of cementitious materials are recommended
variables on the workability and compressive strength of alkali-
by Sika. Recover is a hydration stabilizer and complies with ASTM
activated Class C fly ash. Six admixtures were chosen based on
C494 [26] as a Type D retarder. It is an aqueous solution of hydrox-
results reported in prior literature and included: sodium gluconate,
ycarboxylic acid salts and compound carbohydrates and has
a commercial cement hydration stabilizer, borax, a commercial
sodium gluconate, sucrose, and water as the main ingredients.
naphthalene sulfonate-based admixture, sodium sulfate, and
For traditional applications, dosage rates of 130–390 mL/100 kg
fumed silica. The mixing process plays a key role in workability.
of cement are used. Pretesting of proper dosage rates is recom-
Increasing the intensity of mixing can result in increasing the ini-
mended by the manufacturer.
tial workability state of a cementitious mixture (e.g. a reduced
Aerosil 200 fumed silica (Evonik) was also used as an admix-
yield stress, reduced viscosity, or increased slump/slump flow)
ture. The silica was allowed to dissolve in the activating solution
[17,18]. In alkali-activated fly ash, increasing the mixing time has
longer than the previously mentioned admixtures in order to allow
been correlated with increases in setting times [19]. However, a
for more complete dissolution. The fumed silica was dissolved in
reduction in the initial workability state (e.g. slump), as well as a
the 4 M NaOH solution to achieve a silica modulus (SiO2/Na2O
workability loss, have also been reported with increases in mixing
molar ratio) of 1 (120 g SiO2 in 1 L of solution) after the initial
intensity in cementitious materials [20–22]. Thus, the influence of
24 h equilibration period, and the resulting sodium silicate solu-
the mixing process on the workability and strength of the AAM was
tion was allowed to equilibrate for another 24 h.
also examined. Alkali-activated pastes were examined using
mini-slump and compressive strength testing using three different
2.2. Methods
Class C fly ashes to examine the effect of material variability on
properties.
Most alkali-activated fly ash pastes were prepared using a plan-
etary mixer (Hobart N50). A mass ratio of NaOH solution to fly ash
2. Materials and methods (solution-to-powder ratio) of 0.45 was used for all specimens.
When using the planetary mixer, the mixing procedures followed
2.1. Materials the mixing procedures for pastes as described in ASTM C 305
[27]. For the planetary mixing, the first speed revolved the paddle
Three sources of ASTM C618 [23] Class C fly ash from the three at a rate of 140 ± 5 rpm, with a planetary motion of approximately
different power plants in the United States, referred to as P, B, and 62 rpm. The second speed revolved the paddle at a rate of 285 ± 10
R, were used. The oxide compositions of the fly ashes, provided by rpm, with a planetary motion of approximately 125 rpm.
the supplier, are shown in Table 1. In spite of coming from different For some pastes, an overhead mixer (SCILOGEX OS40-S) was
sources, the fly ashes have quite similar compositions, with cal- used to determine the effect of mixing method on properties. The
cium oxide contents ranging from 24 to 29% and silica to alumina mixing procedures for the overhead mixer were as follows:
228 W. Rakngan et al. / Construction and Building Materials 183 (2018) 226–233

1) Added the fly ash to the NaOH solution and rested for 30 s.
2) Started the mixer and mixed at 350 ± 10 rpm for 60 s.
3) Stopped the mixer for 30 s and scraped the paste collected
on sides of the container.
4) Started the mixer and mixed at 1000 ± 10 rpm for 60 s.

Consistency of the fresh alkali-activated fly ash pastes was eval-

uated by using a mini-slump test procedure [28] using a stainless
steel mini-slump cone. The mini-slump cone had dimensions of
40 mm diameter on the bottom, 20 mm diameter on the top, and
60 mm height. The paste was poured into the cone in a single pour
and then it was tamped 15 times using a spatula. Excess paste from
the top was scraped off, and the cone was lifted gently in the ver-
tical direction. For each measurement, two perpendicular diame-
ters of the patty were measured and the mini-slump area was
calculated from the average value of the diameters. The measure-
ments were performed at 5, 10, 20, 30, 45, and 60 min after mixing.
The paste was kept in a sealed container between measurement
periods, and a new portion of the stored paste was tested at each
time interval, with re-stirring before testing. The test was stopped
when the mini-slump area was less than 1500 mm2 (equivalent
diameter of 43.7 mm), herein defined as the ‘‘workable limit.”
When overhead mixing was used, mini-slump measurements were
conducted at 5, 20, 40, and 60 min after mixing. The ‘‘workable
time” for a mixture was defined as the time until the mini-slump
area reached 1500 mm2.
For compressive strength testing, pastes were cast in 50 mm
diameter and 100 mm height plastic cylinder molds with lids cov-
ering the fresh surfaces. The specimens were left at room temper-
ature for 1 h and cured at 38 °C and 95% humidity for 24 h. They
were then demolded and cured at 23 °C and 100% RH until the time
of testing. During the compressive strength test, neoprene pads
were used to assure load uniformity and the loading rate was con-
trolled within 391–587 N/s. The compressive strength test was
performed at 7 and 28 days after the specimens were cast and
the average of three tests was calculated. In some cases, strength
testing was replicated and these values were combined with the
original test results.

3. Results and discussion

3.1. Effect of admixtures on mini-slump loss

Without chemical admixtures, the mini-slump areas of all

pastes were less than the workable limit, 1500 mm2, at the first
measurement at 5 min, as shown in Fig. 1 (labeled as ‘‘control”).
Therefore, all three fly ashes showed very poor workability in the
absence of admixtures when alkali-activated. This is consistent
with results of other studies [2,3].
Several of the chemical admixtures tested had no effect on
mini-slump area at the dosages tested. These included borax
(sodium tetraborate decahydrate) at 0.5%–10%, Sikament N Fig. 1. Mini-slump loss for alkali-activated fly ash pastes containing sodium
(naphthalene sulfonate) at 1%–10%, and sodium sulfate at 1%–5%. gluconate (SG): a. paste with B-fly ash, b. paste with P-fly ash, c. paste with R-fly
ash.
In all of these pastes the initial slump areas at 5 min were less than
1500 mm2 (data not shown). In addition, qualitatively it appeared
that the higher dosages of borax and Sikament N increased the With 0.25% sodium gluconate addition, the initial mini-slump area
cohesiveness and stickiness of the pastes. No further examinations increased substantially to 9300–13400 mm2, regardless of the fly
were carried out on these mixtures. ash used to prepare the paste, and the workable time also
Fig. 1 shows the results of reagent-grade sodium gluconate increased. With 0.35% sodium gluconate, the initial slump areas
(referred to as SG in the figure) addition (0.1 wt%–0.5 wt%) on were also high for all the fly ash pastes, and the workable times
the mini-slump loss for pastes prepared using the different fly improved to 30–45 min. By increasing the sodium gluconate
ashes. With 0.10% SG addition, the initial mini-slump area dosage to 0.50%, the mini-slump areas at 60 min were at least
(5 min) was the same as the control for the B-fly ash, so the points 1500 mm2 for all fly ashes. For different sources of fly ash with
overlap in Fig. 1a. However, the initial workability was slightly the same dosage of sodium gluconate, generally the slump areas
improved with 0.10% SG for the P and R-fly ashes (Fig. 1b and c). of R-fly ash pastes were greatest.
 W. Rakngan et al. / Construction and Building Materials 183 (2018) 226–233 229

Recover, a commercial admixture containing sodium glu-

conate, was used at dosages of 1 wt%–2 wt% of the fly ashes, with
mini-slump results shown in Fig. 2. Even at the lowest dosage of
1% Recover, the initial slump areas were high, 4800–9800 mm2,
indicating good workability. Increasing the dosage of Recover
increased the workable time from 20 to 30 min with 1.25%
Recover to at least 60 min with 2% Recover, and also slowed
the rate of mini-slump loss. For different sources of the fly ash
with the same dosage of Recover, the slump areas of R fly ash
pastes were greatest, followed by those of P and B-fly ashes,
respectively.

Fig. 3. Effect of adjusting silica modulus on mini-slump loss of alkali-activated fly

ash pastes.

The three fly ashes were also tested using a single dosage of
fumed silica, added to the NaOH solution to obtain a molar ratio
SiO2/Na2O of 1 (silica modulus, Ms = 1). For all fly ashes, the pastes
prepared with 4 M NaOH solution and Ms = 1 exhibited high initial
slump areas of 8900–12,500 mm2 (Fig. 3). In addition, the pastes
sustained high mini-slump areas over 60 min of testing. The P-fly
ash paste had the largest mini-slump area with fumed silica addi-
tion, in contrast to the sodium gluconate and Recover additions
where the R-fly ash pastes had the largest areas.

3.2. Effect of mixing on mini-slump loss

The effect of overhead mixing on mini-slump was only tested

with the B and R-fly ashes; results are shown in Figs. 4 and 5,
respectively. Changing the mixing method for the B-fly ash pastes
did not impact the initial mini-slump areas of the pastes
containing SG or Recover. For example, the initial (5 min) mini-
slump area of the ‘‘B + 0.35%SG:Planetary” sample was the same
as the ‘‘B + 0.35%SG:Overhead” sample in Fig. 4. However, the
admixture-containing pastes prepared using overhead mixing
had increased workable times compared to admixture-containing
pastes prepared using planetary mixing. For example, the workable
times for both the 0.35% sodium gluconate-containing paste and
1.50% Recover-containing paste were greater than 60 min when

Fig. 2. Mini-slump loss for alkali-activated fly ash pastes containing Recover: a. Fig. 4. Effect of overhead mixing on mini-slump loss of alkali activated B-fly ash
paste with B-fly ash, b. paste with P-fly ash, c. paste with R-fly ash. pastes with and without admixtures.
230 W. Rakngan et al. / Construction and Building Materials 183 (2018) 226–233
Fig. 5. Effect of overhead mixing on mini-slump loss of alkali activated R-fly ash pastes with and without admixtures: a. sodium gluconate, b. Recover, c. delayed addition of
sodium gluconate, d. Ms = 1.
they were prepared using overhead mixing, also with a slower rate
of slump loss, compared to approximately 30 min using planetary
mixing.
For R-fly ash pastes, overhead mixing did not change the ini-
tial slump area for control pastes (not shown). The R-fly ash
pastes were tested with a wider range of sodium gluconate
dosages than the B-fly ash pastes (0.35% for B compared to
0.25–0.50% for R). Similar to the B-fly ash pastes with sodium
gluconate, overhead mixing did not change the initial slump
areas in R-fly ash pastes, but did increase mini-slump areas at
later times compared to the pastes prepared using a planetary
mixer (Fig. 5a). The increased mini-slump retention in the over-
head mixed pastes resulted in extending the workable time, in
some cases as much as 40 min (e.g. the ‘‘R + 0.25%SG:Planetary”
paste compared to the ‘‘R + 0.25%SG:Overhead” paste in Fig. 5a).
Overhead mixing had a variable effect on the initial slump area
of R-fly ash pastes containing Recover. At a 1.00% Recover dosage,
overhead mixing resulted in increasing the initial slump area by
about 35%. However, the initial slump areas of pastes prepared
with high dosages of Recover (1.50% and 2.00%) and overhead mix-
ing were lower than their counterpart pastes prepared using plan-
etary mixing. Work conducted in Portland cement paste systems
showed that a decrease in workability can occur in pastes contain-
ing admixtures when mixing intensity increased past a critical
threshold due to changes in pore solution chemistry and the
increased mixing impacting cement-admixture interactions
[22,29]. That the initial slump area with high Recover dosages
was reduced by overhead mixing suggests that an amplification
in aggregation kinetics occurred, possibly due to the combined
effects of increased mixing intensity increasing particle-particle
collisions [30] and changes in the pore solution chemistry
[31–33] due to additional molecules from the Recover admixture
in the system. Similar to the sodium gluconate-containing pastes,
the mini-slump areas at later measurement times increased for
all Recover dosages with overhead mixing, suggesting that the
overhead mixing considerably impacted the rigidity of the fresh
state network (through impacting the cementitious-phase reaction
mechanisms, agglomeration kinetics, and/or N-A-S-H phase forma-
tion) during the time-frame considered in this work. Further work
is needed to elucidate the driving mechanisms governing the
behavior observed. Overhead mixing did not change the mini-
slump results for R-fly ash pastes with a silica modulus of 1
(Fig. 5d), which suggests that the fresh microstructure for the
Ms = 1 pastes was unchanged.
Delayed addition of the admixture was also tested with sodium
gluconate added just before overhead mixing at 1000 rpm for this
test (step 4 in the mixing procedure); this represents a 1.5 min
delay compared to the normal mixing procedure. The delayed
addition of sodium gluconate also did not affect the initial slump
area, but considerably improved the slump flow at other measure-
ment times (Fig. 5c). It is likely that delayed addition of the admix-
ture improved slump retention for in alkali-activated fly ash pastes
for the same reasons it affects slump retention in Portland cement
systems [34,35].
 W. Rakngan et al. / Construction and Building Materials 183 (2018) 226–233
3.3. Effect of admixtures on compressive strength
Fig. 6 shows the effects of sodium gluconate on compressive
strength of pastes made with the three fly ashes. With sodium
gluconate dosages of 0.10% and 0.25%, the compressive strengths
of the pastes were comparable to or higher than the control pastes
at both 7 and 28 days. This increase in the compressive strength
could be partially due to better consolidation of the specimens
due to the improved workability; control paste cylinders had
visible voids on the surface due to inadequate compaction, while
the sodium gluconate-containing paste cylinder surfaces were
smooth.
On the other hand, with sodium gluconate dosages of 0.35% and
0.50%, the compressive strengths of the pastes at 7 and 28 days sig-
nificantly decreased compared to the control pastes. This suggests
that sodium gluconate should not be added at dosages of more
than 0.25% because it can cause significant retardation of reaction
product formation. This is disappointing, since mini-slump reten-
tion was substantially improved at dosages of 0.35% and higher
(Fig. 1).
The effect of Recover on the compressive strength varied with
the source of fly ash, as shown in Fig. 7, even though the fly ashes
have similar compositions (Table 1). For the B-fly ash, a 1% addition
of Recover did not affect the compressive strength. However, for
Fig. 6. Compressive strength of for alkali-activated fly ash pastes containing
sodium gluconate.
Fig. 7. Compressive strength of for alkali-activated fly ash pastes containing
Recover.
231
Fig. 8. Effect of adjusting silica modulus on compressive strength of alkali-activated
fly ash pastes.
the P and R-fly ashes, adding 1% Recover reduced strength. For
all fly ashes, additions of 1.50% Recover reduced strength. Conse-
quently, an acceptable dosage of Recover for the fly ashes should
be lower than 1.50%. Again, this is in conflict with the mini-
slump tests, which suggested that Recover should be used at a
1.50% dosage to obtain good slump retention. As with sodium glu-
conate, this conflict suggests that Recover is not an appropriate
admixture for slump retention because of the retardation in reac-
tion product formation.
For sodium gluconate and Recover additions, pastes made with
R-fly ash tended to have lower compressive strengths than those
made with B and P-fly ashes, and fluctuations in loading rate were
usually encountered during the mechanical testing of the speci-
mens made with R fly ash because the strengths were so low.
Addition of fumed silica to the activating solution to achieve
Ms = 1 significantly affected compressive strength of the pastes
as shown in Fig. 8. For the P and B-fly ashes, the compressive
strengths of the pastes increased significantly by increasing the
silica modulus. However, the strength of the R-fly ash pastes
dropped substantially. For P and B-fly ashes, improvements in
the compressive strengths of the pastes were probably due to bet-
ter consolidation of the specimens and increases in SiO2/Al2O3,
which is known to result in a more homogeneous microstructure
with smaller pores [36]. For R-fly ash, the reason for the strength
decrease is unclear. However, soft surfaces of the paste specimens
were observed during demolding and before the compressive
strength tests. This problem merits further investigation in the
future. The addition of fumed silica to achieve Ms = 1, therefore,
is not a reliable method of improving slump retention for alkali-
activated fly ash.
3.4. Effect of mixing on compressive strength
The effect of mixing method on compressive strength was only
tested for alkali-activated B-fly ash pastes, shown in Fig. 9. Without
admixtures, the compressive strengths at 7 and 28 days of the
pastes prepared by overhead mixing were very similar to those
of the pastes mixed with a planetary mixer. However, mixing
method did result in differences in compressive strengths of pastes
containing admixtures. Overhead mixing doubled the 7-day
strength for the 0.35% sodium gluconate-containing paste and
the 28-day strength of the 1.5% Recover paste. However, the
strengths of the pastes containing admixtures, regardless of mixing
method, were still quite low compared to the control pastes. There-
fore, although overhead mixing extended workable times of pastes,
it could not combat the strength losses from admixture use.
232 W. Rakngan et al. / Construction and Building Materials 183 (2018) 226–233
under the Exploratory Advanced Research Program, award number
DTFH61-13-H-00011. W. Rakngan would also like to acknowledge
support for his graduate studies from SCG.
Fig. 9. Effect of overhead mixing on compressive strength of alkali activated B-fly
ash pastes with and without admixtures.
4. Conclusions
Several chemical admixtures were examined for their ability to
improve workability of alkali-activated Class C fly ash pastes
through the measurement of mini-slump retention over time.
Some admixtures that had been suggested in the literature to
improve workability had no impact on the workability of the
alkali-activated Class C fly ash pastes in this study, including:
borax, sodium sulfate, and a commercial naphthalene sulfonate-
based water reducer. Sodium gluconate, which is a known retarder
of Portland cement hydration, and a commercial hydration stabi-
lizer containing sodium gluconate were both effective at improving
initial paste workability as well as mini-slump retention, with
effects improving as the admixture dosage was increased. How-
ever, with these admixtures, the dosage necessary to maintain ade-
quate workability over time caused a dramatic loss in paste
compressive strength at 7 and 28 days, indicating that these are
not appropriate admixtures for this system. Modifying the silica
modulus of the activating solution to a value of Ms = 1 provided
excellent initial workability and retention, with pastes maintaining
workability to at least 60 min. In fact, with two of the fly ashes, an
activating solution with Ms = 1 increased paste compressive
strength. In the fly ash whose pastes had the lowest 7-day strength
for the control mixture, however, an activating solution of Ms = 1
actually decreased the strength even further.
Overhead mixing on its own was not able to modify the initial
workability of the pastes, but did considerably contribute to
improved workability retention in pastes containing retarders.
Slump retention was even better with a delayed addition of the
admixture. However, the overhead mixing was not able to overcome
the negative effects of the admixtures on compressive strength.
From this study, it appears that the most promising method of
controlling workability of alkali-activated Class C fly ash is through
the manipulation of the silica modulus of the activating solution.
Further work is needed to optimize the solution composition and
understand the role of the solution compressive strength, because
the effect on strength is not predictable.
Conflict of interest
None.
Acknowledgements
The authors would like to acknowledge funding for this
research from the United States Federal Highway Administration
References
[1] RILEM TC 224-AAM, Alkali Activated Materials, State-of-the-Art Report,
RILEM, 2014.
[2] E.I. Diaz-Loya, E.N. Allouche, S. Vaidya, Mechanical properties of fly-ash-based
geopolymer concrete, ACI Mater. J. 108 (2011) 300–306.
[3] P. Chindaprasirt, P. De Silva, K. Sagoe-Crentsil, S. Hanjitsuwan, Effect of SiO2
and Al2O3 on the setting and hardening of high calcium fly ash-based
geopolymer systems, J. Mater. Sci. 47 (2012) 4876–4883.
[4] X. Guo, H. Shi, W.A. Dick, Compressive strength and microstructural
characteristics of class C fly ash geopolymer, Cem. Concr. Compos. 32 (2010)
142–147.
[5] J.E. Oh, P.J.M. Monteiro, S.S. Jun, S. Choi, S.M. Clark, The evolution of strength
and crystalline phases for alkali-activated ground blast furnace slag and fly
ash-based geopolymers, Cem. Concr. Res. 40 (2010) 189–196.
[6] F. Winnefeld, A. Leemann, M. Lucuk, P. Svoboda, M. Neuroth, Assessment of
phase formation in alkali activated low and high calcium fly ashes in building
materials, Constr. Build. Mater. 24 (2010) 1086–1093.
[7] M. Palacios, F. Puertas, Stability of superplasticizer and shrinkage-reducing
admixtures in high basic media, Mater. Constr. 54 (2004) 65–86.
[8] E. Douglas, J. Brandstetr, A preliminary study on the alkali activation of ground
granulated blast-furnace slag, Cem. Concr. Res. 20 (1990) 746–756.
[9] A. Sathonsaowaphak, P. Chindaprasirt, K. Pimraksa, Workability and strength
of lignite bottom ash geopolymer mortar, J. Hazard. Mater. 168 (2009)
44–50.
[10] P. Chindaprasirt, T. Chareerat, V. Sirivivatnanon, Workability and strength of
coarse high calcium fly ash geopolymer, Cem. Concr. Compos. 29 (2007) 224–
229.
[11] M. Palacios, Y.F. Houst, P. Bowen, F. Puertas, Adsorption of superplasticizer
admixtures on alkali-activated slag paste, Cem. Concr. Res. 39 (2009) 670–677.
[12] B. Nematollahi, J. Sanjayan, Effect of different superplasticizers and activator
combinations on workability and strength of fly ash based geopolymer, Mater.
Des. 57 (2014) 667–672.
[13] F. Collins, J.G. Sanjayan, Early age strength and workability of slag pastes
activated by NaOH and Na2CO3, Cem. Concr. Res. 28 (1998) 655–664.
[14] U. Rattanasak, K. Pankhet, P. Chindaprasirt, Effect of chemical admixtures on
properties of high-calcium fly ash geopolymer, Int. J. Min. Met. Mater. 18
(2011) 364–369.
[15] Antoni, S.W. Wijaya, J. Satria, A. Sugiarto, D. Hardjito, The use of borax in
deterring flash setting of high calcium fly ash based geopolymer, Mater. Sci.
Forum 857 (2016) 416–420.
[16] M.J. Nadoushan, A.A. Ramezanianpour, The effect of type and concentration of
activators on flowability and compressive strength of natural pozzolan and
slag-based geopolymers, Constr. Build. Mater. 111 (2016) 337–347.
[17] D.A. Williams, A.W. Saak, H.M. Jennings, Influence of mixing on the rheology of
fresh cement paste, Cem. Concr. Res. 29 (1999) 1491–1496.
[18] D. Lowke, P. Schiessl, Effect of mixing energy on fresh properties of SCC, SCC
2005, Proceedings of the 4th International RILEM Symposium on Self-
Compacting Concrete, Hanley Wood Publication, Chicago, IL, 2005.
[19] P. Chindaprasirt, P. De Silva, S. Hanjitsuwan, Influence of high-speed mixing on
properties of high calcium fly ash geopolymer, Arab. J. Sci. Eng. 39 (2014)
6001–6007.
[20] J. Dils, G. De Schutter, V. Boel, Influence of mixing procedure and mixer type on
fresh and hardened properties of concrete: a review, Mater. Struct. 45 (2012)
1673–1683.
[21] R. Helmuth, Abnormal Concrete Performance in the Presence of Admixtures,
PCA R&D Serial No. 2006, Portland Cement Association, 1995.
[22] D. Han, R. Ferron, Effect of mixing method on microstructure and rheology of
cement paste, Constr. Build. Mater. 93 (2015) 278–288.
[23] ASTM C618-15, Standard Specification for Coal Fly Ash and Raw or Calcined
Natural Pozzolan for Use in Concrete, ASTM International, West
Conshohocken, PA, 2015. www.astm.org.
[24] T. Williamson, M.C.G. Juenger, The role of activating solution concentration on
alkali silica reaction in alkali-activated fly ash concrete, Cem. Concr. Res. 83
(2016) 124–130.
[25] S. Hanjitsuwan, S. Hunpratu, P. Thongbai, S. Maensiri, V. Sata, P. Chindaprasirt,
Effects of NaOH concentrations on physical and electrical properties of high
calcium fly ash geopolymer paste, Cem. Concr. Compos. 45 (2014) 9–14.
[26] ASTM C494/C494M-16, Standard Specification for Chemical Admixtures for
Concrete, ASTM International, West Conshohocken, PA, 2016. www.astm.org.
[27] ASTM C305-14, Standard Practice for Mechanical Mixing of Hydraulic Cement
Pastes and Mortars of Plastic Consistency, ASTM International, West
Conshohocken, PA, 2014. www.astm.org.
[28] D.L. Kantro, Influence of water-reducing admixtures on properties of cement
paste – a miniature slump test, Cem. Concr. Aggr. 2 (1980) 95–102.
[29] D. Han, R. Ferron, Influence of high mixing intensity on rheology, hydration
and microstructure of fresh state cement paste, Cem. Concr. Res. 84 (2016) 95–
106.
[30] R. Ferron, S. Shah, E. Fuente, C. Negro, Aggregation and breakage kinetics of
fresh cement paste, Cem. Concr. Res. 50 (2013) 1–10.
 W. Rakngan et al. / Construction and Building Materials 183 (2018) 226–233
[31] M. Yang, C.M. Neubauer, H.M. Jennings, Interparticle potential and
sedimentation behavior of cement suspensions: review and results from
paste, Adv. Cem. Based Mater. 5 (1997) 1–7.
[32] H. Tan, F. Zou, B. Ma, Y. Guo, X. Li, J. Mei, Effect of competitive adsorption
between sodium gluconate and polycarboxylate superplasticizer on rheology
of cement paste, Constr. Build. Mater. 144 (2017) 338–346.
[33] M. Zajac, J. Skocek, F. Bullerjahn, M.B. Haha, Effect of retarders on the early
hydration of calcium-sulpho-aluminate (CSA) type cements, Cem. Concr. Res.
84 (2016) 62–75.
233
[34] H. von Daake, D. Stephan, Impact of retarders by controlled addition on the
setting, early hydration and microstructural development of different
cements, Mag. Concr. Res. 68 (2016) 1011–1024.
[35] J. Cheung, A. Jeknavorian, L. Roberts, D. Silva, Impact of admixtures on the
hydration kinetics of Portland cement, Cem. Concr. Res. 41 (2011) 1289–1309.
[36] P. Duxson, A. Fernández-Jiménez, J.L. Provis, G.C. Lukey, A. Palomo, J.S.J. van
Deventer, Geopolymer technology: the current state of the art, J. Mater. Sci.
(2007) 2917–2933.