materials
Article
Investigating the Impact of Fly Ash on the Strength and
Micro-Structure of Concrete during Steam Curing and
Subsequent Stages
Yun Duan 1 , Qicai Wang 1,2, *, Zhaofei Long 1 and Xiaoping Wang 1
Citation: Duan, Y.; Wang, Q.; Long,
Z.; Wang, X. Investigating the Impact
of Fly Ash on the Strength and
Micro-Structure of Concrete during
Steam Curing and Subsequent Stages.
Materials 2023, 16, 1326. https://
doi.org/10.3390/ma16041326
Academic Editor: Francesco
Fabbrocino
Received: 8 January 2023
Revised: 20 January 2023
Accepted: 24 January 2023
Published: 4 February 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Materials 2023, 16, 1326. https://doi.org/10.3390/ma16041326
1 Civil Engineering Department, Lanzhou Jiaotong University, Lanzhou 730070, China
2 National and Provincial Joint Engineering Laboratory of Road & Bridge Disaster Prevention and Control,
Lanzhou Jiaotong University, Lanzhou 730070, China
* Correspondence: wqaici62@163.com
Abstract: Steam curing, a common way of curing precast concrete, can greatly improve its 1-day
strength. However, the thermal effect of prolonged high-temperature curing can negatively impact
the concrete’s performance, thus compromising production of precast products in cold regions. Fly
ash (FA) is used as a supplementary cementitious material to improve part of the properties of
concrete. In this paper, we investigated the effect of FA (10~30%) on the compressive strength and
microstructure of manufactured sand concrete at the steam curing and later stages. Specifically, we
analyzed the behavior of FA in the constant temperature phase under steam curing. Results indicated
that the pozzolanic reaction of FA started to occur at 24 h of constant temperature curing. Early
hydration under steam curing produces a large amount of Ca(OH)2 , causing the pozzolanic reaction
of FA to occur significantly earlier, and the high pH value of the solution and the fibrous mesh
structure of the FA surface promote the pozzolanic reaction. The addition of 30% FA to manufactured
sand concrete causes a significant reduction in early strength under steam curing, which is not
beneficial to the formwork removal and tensioning of precast members. Notably, manufactured sand
concrete with 20% FA under steam curing had the highest late strength. The filling effect of FA and
the additional gel produced by the pozzolanic reaction would result in the reduction in large pore
content, refinement of pore size, improvement of microstructural compactness, and increase in gel
system strength. Therefore, the addition of 20% FA to the manufactured sand concrete can improve
the long-term strength, which is beneficial to the production of precast beams in cold regions.
Keywords: fly ash; compressive strength; microstructure; steam curing; manufactured sand concrete;
pozzolanic reaction
1. Introduction
Precast concrete products play an important role in infrastructural development. The
sleepers and precast beams used in China‘s high-speed railway are precast prestressed
concrete components [1,2]. To improve the productivity of precast products, steam curing
is often used to shorten the time required to reach demoulding and prestress tensioning
strength [3–5]. Typically, the temperature of steam curing of concrete is between 50 ◦ C
and 70 ◦ C [6,7], but higher than 70 ◦ C after curing. Studies have shown that delayed
ettringite formation (DEF) not only causes volume expansion, but also leads to concrete
cracking [8–10]. Therefore, increasing curing temperature rarely improves the early strength
of precast girder plants. The duration of the constant temperature phase in steam curing
is usually controlled at 4–12 h, and the strength at the end of steam curing can generally
meet the demolding strength (30 MPa) [1,4,11]. However, it is difficult to achieve the
requirements for prestress tensioning strength (50 MPa) [4,12]. Although elongating the
duration of constant temperature under steam curing can make the strength at the end of
steam curing meet the prestress tensioning requirements, it has also been shown to cause an
https://www.mdpi.com/journal/materials
Materials 2023, 16, 1326 2 of 16

uneven distribution of hydration products and structural looseness, and increase porosity
and large pore content [13–16]. Consequently, these phenomena have been shown to reduce
concrete’s long-term strength, thereby seriously affecting the performance and service life
of concrete structures [6,11,17,18]. Winter in the north of China is cold and lasts a long time.
For precast components produced under this environment, engineers expect the strength
of the concrete at the end of steam curing to meet prestress tensioning requirements. This
is to both increase the productivity of the precast plant and reduce production costs and
energy consumption by eliminating subsequent curing.
Reducing the harmful effects of prolonged steam curing on concrete properties and
microstructure is imperative to achieving the engineers’ goals. Fly ash (FA), a solid waste
from coal burning, is widely used in concrete as a typical supplementary cementitious
material [19,20]. Partial replacement of cement by FA not only makes wise use of solid
waste resources, but also reduces carbon dioxide emissions and energy consumption [21].
Externally, FA particles appear spherical, a phenomenon that facilitates dispersion of
cement particles and promotes the sliding of neighboring particles, while the ball-bearing
effect enhances the fluidity of fresh concrete [22–24]. FA has a particle size that ranges
between 1–100 µm. FA‘s microspheres provide nucleation sites for C-S-H gels at the early
stage of hydration, which promotes the hydration of the cement [25,26]. The inert FA acts
as a filler to reduce the large pore content [19,27]. Glassy FA contains reactive SiO2 and
Al2 O3 , which can react with calcium hydroxide (CH) produced by cement hydration to
produce more calcium silicate hydrate (C-S-H), calcium aluminate hydrate (C-A-H) and
calcium aluminum silicate hydrate (C-A-S-H). These gels have been shown to improve the
late strength and reduce the capillary content of concrete [28–30]. Analysis of FA‘s positive
effects indicates that they have the potential to reduce the negative effects of the thermal
effects of steam curing. Therefore, it is necessary to evaluate the behavior of FA under
steam curing.
FA‘s low reactivity, coupled with its delayed effect on setting time [20,26], makes it less
than 35% of mass of cement replacement in concrete [19,31]. Using steam curing for FA con-
crete not only improves FA‘s reactivity, increases the concrete‘s early strength [1,32–34] and
completes the demoulding of precast products as early as possible, but also increases the
amount of FA used [31]. Yazıcı et al. [35] studied the compressive strength of concrete with
different FA (class C) contents after 5 h of constant temperature (65 ◦ C) under steam curing,
and found that the 1-day strength decreased with the increase in FA content. However, after
the substitution of FA exceeded 30%, the concrete had a strength less than 30 MPa for 1 day,
which was not conducive to the removal of precast formwork. The strength first increased
at 28 days, then decreased with increasing FA content. During this process, concrete with
20% FA had the greatest strength. However, the authors did not explain this phenomenon.
Jiang et al. [36] showed that the late strength of concrete with 30% FA (class F) was lower
than that of pure cement at a constant temperature (60 ◦ C) of 8 h under steam curing. With
the FA content increased, the late strength was reduced. Previous studies have shown that
high-temperature curing accelerates the pozzolanic reaction of FA, which causes a release
of Al3+ and Si4+ into the C-S-H gel structure and connected silicon short chains, thereby
increasing the content of silicate tetrahedra coordinated with aluminate tetrahedra, and
increasing in the polymerization of C-S-H gels [37]. Chindaprasirt et al. [38] found that the
28d compressive strength of FA (class C) concrete was higher than that of cement concrete.
As mentioned above, although much research has been conducted in this field, several
limitations still exist. For example, controlling constant temperature duration in the current
steam curing regime does not effectively meet the tensioning requirements of precast beams,
especially for the production of precast beams in cold regions. Studies on FA concrete
have mainly focused on normal temperature curing, and there is a lack of research on
the strength and microstructure of FA concrete under steam curing, especially the steam
curing process, because it is a critical period for the formation of concrete properties and
microstructure. It is certain that high temperature curing can have harmful effects on the
 (XRD) analysis, scanning electron microscopy (SEM) and nuclear m
(NMR) techniques to investigate the properties, microstructure and p
hydration products and FA’s pozzolanic reaction products under stea
Materials 2023, 16, 1326 3 of 16
2. Materials and Methods
2.1. Materials
long-term strength and microstructure of concrete, but only a few studies have explored
how to The
reducePortland
the thermalcement (P Ⅱ 52.5) produced in Lanzhou Qilianshan
damage effect.
In this paper, we targeted production of precast beams in cold regions with the aim of
a specific surface of 351 m2/kg, meets the requirements of GB 175-20
understanding the compressive strength and microstructure of FA concrete at the constant
of FA was
temperature generated
stage by burning
of steam curing. bituminous
Next, we discuss the effectcoal,
of FA which was obtaine
on the long-term
strength of steam-cured manufactured sand concrete. We
Plant, China. The chemical composition of cement and fly ash used X-ray diffraction (XRD)
was sh
analysis, scanning electron microscopy (SEM) and nuclear magnetic resonance (NMR)
externaltocharacteristics
techniques of FAmicrostructure
investigate the properties, particles and
were poreobserved viahydration
structure of the SEM, while
position
products andwas analyzed
FA’s pozzolanic by energy
reaction productsdispersive X-ray spectroscopy (EDS)
under steam curing.

2.1.Materials
FA‘s phase composition was measured by XRD analysis (Figure
and Methods
size
2.1. distributions of cement and FA were measured by laser diffrac
Materials
The Portland cement
respectively. Results(P II 52.5)
were produced
shown in Lanzhou Qilianshan
in Figure Cement
3. FA‘s D50Plant,value
with a was s
specific surface of 351 m2/kg, meets the requirements of GB 175-2007 [39]. Class F type of
cement.
FA The by
was generated coarse
burningaggregate
bituminous coal,was 5–20
which wasmm particle-sized
obtained from Xigu Powercrushed
Plant, st
der content
China. The chemicaland 12.4% of
composition crushing
cement andindex. The
fly ash was fine
shown aggregate
in Table was limes
1. The external
characteristics of FA particles were observed via SEM, while their chemical composition was
sand with
analyzed a fineness
by energy modulus
dispersive X-ray of 2.91,
spectroscopy (EDS),aaspowder content
shown in Figure 1. FA‘sof 5.6% an
phase
value of was
composition 0.8%. All these
measured by XRD met the
analysis requirements
(Figure of GB/T
2), whereas particle 14684 [40]. T
size distributions
of cement and FA were measured by laser diffraction granulometry, respectively. Results
superplasticizer with solids content of 23.6% and water reduction rate
were shown in Figure 3. FA‘s D50 value was smaller than that of cement. The coarse
improve
aggregate wasthe
5–20 workability
mm particle-sizedof the stone
crushed manufactured
with 0.4% powdersand concrete
content and 12.4%and ac
crushing index. The fine
slump requirements [41]. aggregate was limestone manufactured sand with a fineness
modulus of 2.91, a powder content of 5.6% and a methylene blue value of 0.8%. All these
met the requirements of GB/T 14684 [40]. The polycarboxylate superplasticizer with solids
Tableof1.23.6%
content Chemical compositions
and water reduction rate of Portland
of 27% was usedcement andthe
to improve FA.workability of
the manufactured sand concrete and achieve the expected slump requirements [41].
Chemical
Table 1. Chemical compositions of Portland cement and FA.
Compositions SiO2 CaO Al2O3 FexOy K2O MgO Na2O SO3 T
Chemical Compositions Loss of
(% by Mass)
SiO2 (%
CaOby Mass)
Al2 O3 Fex Oy K2 O MgO Na2 O SO3 TiO2
Ignition
PC 22.06 61.56 PC4.45 3.46 22.06
0.6861.56
2.544.450.313.46 2.67
0.68 0.37
2.54 1.88
0.31 2.67
FA 54.92 5.91 30.76 3.07 1.14 0.88 0.51 0.08 - 2.56
FA 54.92 5.91 30.76 3.07 1.14 0.88 0.51 0.08

Figure
Figure1. SEM and EDS
1. SEM andof FA.
EDS of FA.
Materials 2023,
Materials 16,16,
2023, x 1326
FOR PEER REVIEW 4 of 16 4 of 16

Figure 2.2.XRD
Figure XRDofofFA.
FA.

Figure
Figure 3.3.Particle
Particlesize
size distributions
distributions of OPC
of OPC and and
FA. FA.

2.2. Concrete Mixing and Specimen Preparation

2.2. Concrete Mixing and Specimen Preparation
Normally, a maximum of 35% by mass of FA can be used to replace cement in con-
crete Normally, a maximum
[19,31], so the admixture of ofFA
35%
by by mass
mass of concrete
in the FA can mixes
be used to replace
of this study wascement
10%, in con-
crete [19,31], so the admixture of FA by mass in the concrete mixes of this study
20% and 30%. Details of the mixing proportions of the concrete with different FA contents was 10%,
10%
20% and 30%.
are shown Details
in Table of mix
2. This the proportion
mixing proportions
(F0) was theofcommon
the concrete with
ratio for thedifferent FA contents
precast beams
are shown
of the in Table
high-speed 2. This
railway mix[41].
in China proportion (F0) was the common ratio for the precast
beams of the high-speed railway in China [41].
Table 2. Mix proportion of concrete with different FA contents.
Table 2. Mix proportion of concrete with different FA contents.
Mix Proportion (kg/m3 )
No.
Mix Proportion
Water (kg/m3) Fly Ash
Cement
Crushed Manufactured
Superplasticizer
Stone Sand
No. Manufactured Superplasticiz
F0 Water 157 Cement
490 Fly
0 Ash Crushed
1148 Stone704 5.3
Sand er
F10 157 441 49 1148 704 5.3
F0
F20 157 157 490
392 98 0 1148 1148 704 704 5.3 5.3
F30
F10 157 157 343
441 147 49 1148 1148 704 704 5.3 5.3
F20 157 392 98 1148 704 5.3
◦
F30All concrete
157 specimens
343 were prepared
147 and tested
1148 704 (20 ± 2 C).
at room temperature 5.3
Briefly, materials were first weighed, according to the proportions of the mixture, then the
crushed stone, manufactured sand, cement and fly ash were poured into the forced mixer
All concrete specimens were prepared and tested at room temperature (20 ± 2 °C)
in turn and mixed for 90s. Next, water and polycarboxylate superplasticizer were added
Briefly,
into the materials were
mixture and thefirst weighed,
mixing according
continued to the
for 2 min. Theproportions of the
slump of fresh mixture,
concrete wasthen the
crushed stone,
160–190 mm, manufactured
which sand, cement
met the production and flyofash
requirements werebeams
precast pouredin into the forced
the plant. The mixer
in
fresh concrete was loaded into a cube mold, measuring 100 mm × 100 mm × 100 mm, for added
turn and mixed for 90s. Next, water and polycarboxylate superplasticizer were
into the mixture
compressive andand
strength theNMR
mixing continued
testing. for 2 min.
The concrete The slump
specimens loaded of fresh
into concrete was
the molds
160–190 mm, which met the production requirements of precast beams in the plant. The
fresh concrete was loaded into a cube mold, measuring 100 mm × 100 mm × 100 mm, for
compressive strength and NMR testing. The concrete specimens loaded into the molds
were placed on a vibrating table and compacted to reduce air bubbles. The specimens
Materials
Materials 2023,
2023, 16,
16, xx FOR
FOR PEER
PEER REVIEW
REVIEW 55 of
of 11

Materials 2023, 16, 1326 5 of 16

were
were then
then subjected
subjected toto both
both steam
steam curing
curing and
and standard
standard curing.
curing. SEM
SEM specimens
specimens were
were pre
pre
pared
pared by
by selecting
selecting the
the mortar
mortar portion
portion of
of fresh
fresh concrete
concrete to
to reduce
reduce the
the negative
negative
were placed on a vibrating table and compacted to reduce air bubbles. The specimens were effects
effects oo
the
then concrete
the concrete crushing
subjected crushing sampling
sampling
to both steam process
process
curing and on the
on the
standard microstructure
microstructure
curing. of the
of the
SEM specimens cement
werecement matrix.
matrix.
prepared It
by It useuse
the
the same
samethe
selecting curing
curing
mortarregime
portionas
regime asofconcrete.
concrete.
fresh concrete to reduce the negative effects of the concrete
crushing sampling process on the microstructure of the cement matrix. It uses the same
2.3.
2.3. Curing
curing regime
Curing Regimes
as concrete.
Regimes
The
The steam
2.3. Curingsteam curing
curing regime
Regimes regime commonly
commonly used
used for
for the
the production
production of
of precast
precast beams
beams for
for the
th
high-speed
high-speed railway
railway in
in China
China [12]
[12] was used
wasusedused in this study.
study. Curing was completed at a con
The steam curing regime commonly forinthe
thisproduction Curing wasbeams
of precast completed
for the at a con
stant
stant temperature
railway inof
temperature
high-speed of 50
50 °C,
China [12]with
°C, with other
other
was used incontrol
control parameters
this study. Curing wasshown
parameters shown in
completedin Figure
Figure 4.
4. To
at a constant To investi
investi
gate the effect
gate the effect
temperature of ◦
of 50of C,FA
FAwithin the
in the constant
constant
other temperature
control temperature
parameters shown stage of steam
stageinofFigure curing,
steam4.curing, the experimen
the experimen
To investigate
was
the controlled
waseffect
controlled
of FA in by
bythethe
the constant
constant
constant temperature
temperature
temperature curing
stagecuring time,
time,
of steam using
usingthe
curing, the A
A steam
steam curing
theexperiment was regime
curing regim
(Figure
controlled 4).byThethe test time
constant was 6 h,
temperature 12 h, 18
curing h, 24
time,h, 30 h
using and
the 36
A
(Figure 4). The test time was 6 h, 12 h, 18 h, 24 h, 30 h and 36 h of constant temperature h of
steam constant
curing temperature
regime
(Figure
To 4). The
To analyze
analyze thetest
the timeof
effect
effect ofwas
FA6on
FA h, long-term
on 12 h, 18 h, 24
long-term h, 30 h and 36
performance
performance and
andh of constant temperature.
microstructure
microstructure of
of steam-cured
steam-cured
To analyzethe
concrete, the B effect of FA on long-term performance and microstructure of steam-cured
concrete, the B steam curing regime (Figure 4) was used. The steam curing equipment was
steam curing regime (Figure 4) was used. The steam curing equipment wa
concrete,
shown the B steam curing theregime (Figure 4) was used.the The steam curing equipment was
shown in in Figure
Figure 5. 5. After
After the end end of of steam
steam curing,
curing, the concrete
concrete waswas continued
continued to to be
be cured
cured
shown in Figure 5. After the end of steam curing, the concrete was continued to be cured
under standard
understandard
standard curing
curing conditions
conditions until
until the test age.
age. Standard curing was used
used as as aa refer
under curing conditions until the testthe
age.test
Standard Standard
curing was curing
used aswas
a reference refer
ence group,
ence group,
group, with testing
with completed
with testing completed
testing completed
for 28 andforfor 28
9028 and
andThe
days. 90 days.
90 standardThe standard
days. Thecuring
standard curing temperature
curing temperatur
temperature was
was
was
20 ± 220
20◦ C±±and
22 °
°C and
and relative
Crelative relative
humidity humidity
was 96%was
humidity was 96%
96% or
or more. or more.
more.

Figure 4.Steam
Figure4.4.
Figure Steamcuring
Steam regime.
curing
curing regime.
regime.

Figure
Figure5.5.
Figure Steam
5.Steam
Steam curing
curing
curing equipment.
equipment.
equipment.

2.4.
2.4. Methods
2.4. Methods
Methods
2.4.1. Compressive Strength
2.4.1.
2.4.1. Compressive
Compressive Strength
Strength
The compressive strength of concrete was tested according to the China standard
GB/TTheThe compressive
compressive
50081 strength
rate ofof
strength
[42] at a loading 0.6concrete
of concrete was
was tested
MPa/s. Three tested according
according
specimens to
to the
were testedthe China
eachstandard
China
under standard
GB/T
age 50081
of every
GB/T [42]
[42] at
at athen
50081mixture, a loading rate
rate of
an average
loading 0.6
value
of MPa/s.
0.6of Three
Three specimens
the compressive
MPa/s. strength were
specimens was tested
tested under
under each
werecalculated. each
age of every mixture, then an average value of the compressive strength was
age of every mixture, then an average value of the compressive strength was calculated. calculated.
Materials 2023, 16, 1326 6 of 16

2.4.2. Scanning Electron Microscopy

The concrete microstructure was observed by Scanning electron microscopy (SEM),
with a Zeiss Gemini SEM 500. Briefly, test specimens were obtained from mortar specimens
using a small cutting machine, then a plane surface without any grinding treatment was
chosen as the test area. To obtain clear microstructure photos, the specimens were sprayed
with gold to increase their electrical conductivity.

2.4.3. X-ray Diffraction

X-ray diffraction (XRD) method was used to analyze the phase composition of the
specimen. The specimen was taken from the slurry portion of the crushed concrete and
ground into a fine powder The test equipment was a Rigaku Miniflex 600 with an X-ray
source of 60 V. The scan speed was 2◦ /min and the 2-theta range was 5◦ to 65◦ .

2.4.4. Pore Structure

Nuclear magnetic resonance techniques have been widely used to test the pore water
content, pore structure and pore size distribution of cement and concrete materials [43,44].
Based on quantum mechanical property, hydrogen nuclei (1 H) always spin around their
own axis at a certain frequency, which is equivalent to a small constant magnetic field.
In the presence of an external fixed main magnetic field, part of the hydrogen nuclei in
the field is magnetized, which causes them to be distributed parallel to the direction of
the main magnetic field. By applying a set of radio frequency pulses of characteristic
frequencies to the main magnetic field, the low-energy hydrogen protons gain energy to
enter the high-energy state. Consequently, the macroscopic magnetization is deflected and
starts precession. Free induction decay (90-FID) and spin-echo pulse sequence (CPMG)
are the two commonly used pulse sequences. After turning off the radio frequency pulse,
the macroscopic longitudinal magnetization vector gradually returns to the equilibrium
position before excitation under the effect of the main magnetic field. This process becomes
longitudinal relaxation, also called spin–lattice relaxation. The macroscopic transverse
magnetization vector gradually decays, a process called transverse relaxation or spin–spin
relaxation. The time of lattice relaxation is T1 and the time of spin relaxation is T2 [45–47].
T2 can be calculated according to [46] as follows:

1 1 1 1
= + + (1)
T2 T2,bulk T2,surface T2,diffusion

where T2 , bulk is the bulk relaxation time. Free water has a large bulk relaxation time,
about seconds, and thus can be neglected. T2 , surface is the surface relaxation time. T20
diffusion is the diffusion relaxation time, which can also be neglected when the magnetic
field gradient is constant. Therefore, transverse relaxation time T2 can be approximated by
T2 , surface. The relationship between surface relaxation time with pore volume (V) and
pore surface area (S) is calculated as follows:

1 S
=ρ (2)
T2,surface V

where ρ is the material surface relaxivity and takes the value of 12.5 nm/ms [48]. Based on
the dimensional analysis, Equation (2) can be expressed as follows:

1 1
= ρk (3)
T2,surface r

where r and k denote the pore radius and geometric factor of the pore, respectively. In
this study, we assumed that the pore was columnar pore, and the value of k is 2. From
Equation (3), it is evident that the transverse relaxation time of pore water is positively
correlated with the pore size.
 study, we assumed that the pore was columnar pore, and the value of k is 2. From Equa
tion (3), it is evident that the transverse relaxation time of pore water is positively corre
lated with the pore size.
Testing of transverse relaxation time was conducted on a Macro MR12-150H-I (Su
Materials 2023, 16, 1326 zhou, China), with a constant magnetic field of 0.3 ± 0.05 T, a magnet temperature 7 of 16 of 32 ±
0.02 °C and a radio frequency coil diameter of 150 mm. The equipment working frequency
was 12.638 MHz, and can test the size of samples that are less than 150 mm [43,49]. When
Testing ofreached
the samples transverse relaxation
the test agetime wasbefore,
1 day conducted onwere
they a Macro MR12-150H-I
removed from (Suzhou,
the molds, vac
China), with a constant magnetic field of 0.3 ± 0.05 T,
uum-saturated with water for 24 h, and then tested by NMR.a magnet temperature of 32 ± 0.02 ◦ C
and a radio frequency coil diameter of 150 mm. The equipment working frequency was
12.638 MHz, and can test the size of samples that are less than 150 mm [43,49]. When the
3. Results and Discussion
samples reached the test age 1 day before, they were removed from the molds, vacuum-
3.1. Compressive
saturated Strength
with water for 24 h, and then tested by NMR.
The compressive strength of manufactured sand concrete with different FA contents
3. Results and Discussion
under the constant temperature stage of steam curing is shown in Figure 6. The strength
3.1. Compressive Strength
of F0, F10 and F20 concrete could reach the demoulding strength requirement after 6 h o
The compressive strength of manufactured sand concrete with different FA contents
constant temperature curing, while F30 concrete needed 12 h. The strength of F0, F10 and
under the constant temperature stage of steam curing is shown in Figure 6. The strength
F20
of F0,concrete
F10 and could reach could
F20 concrete the prestressing tensioningstrength
reach the demoulding strength requirement
requirement afterafter
6 h 24 h o
constant
of constanttemperature
temperature curing, whileF30
curing, while F30concrete
still could
needed not12reach
h. Thethe prestressing
strength of F0, F10tensioning
strength after 36 h of constant temperature curing. These results
and F20 concrete could reach the prestressing tensioning strength requirement after 24 h indicate that replacing
part
of of thetemperature
constant cement with FA can
curing, whilesignificantly
F30 still couldextend
not reachthethe
steam curing tensioning
prestressing time required for
strength
concreteafter 36 h of
to reach constant
the temperature
demoulding curing.
strength andThese results indicate
prestressing tensionthat replacing
strength. This time
part of the cement with FA can significantly extend the steam curing
can be extended with increase in FA content. All FA concretes under the constant temper time required for
concrete to reach
ature stage the demoulding
of steam curing had strength
lowerand prestressingstrengths
compressive tension strength.
than thoseThis time can cemen
of pure
be extended with increase in FA content. All FA concretes under the constant temperature
concrete under the same age. The FA admixture was increased from 10% to 30% and the
stage of steam curing had lower compressive strengths than those of pure cement concrete
strength
under the of FAage.
same concrete
The FAwas continuously
admixture reduced.
was increased fromThe10%strength
to 30% and of FA concreteofwas neg
the strength
atively linearly correlated with the FA contents before 24
FA concrete was continuously reduced. The strength of FA concrete was negativelyh of constant temperature
linearly cur
ing, with with
correlated the fitting coefficient
the FA contents beforeR 24
2 above 0.85; but
h of constant after 24 h,curing,
temperature R decreased
2 to below 0.7
with the fitting
Notably, the
coefficient R2 above
greater thebut
0.85; amount
after 24 of h,
FA R2replacement,
decreased to belowthe higher the degree
0.7. Notably, the of reduction in
greater
the amount of FA replacement, the higher the degree of reduction in early
early strength under steam curing. The strength difference between FA concrete and pure strength under
steam
cement curing. The strength
concrete difference
was reduced afterbetween FA concrete
24 h of constant and pure cement
temperature curing. concrete was XRD
Subsequent
reduced
SEM and after 24 h of constant
T2 spectra analysistemperature
revealed that curing. Subsequent
the reduced XRD, of
content SEM and T2 spectra
hydration products and
analysis revealed that the reduced content of hydration products and increased total pore
increased total pore content were the main reasons for the low strength of FA concrete
content were the main reasons for the low strength of FA concrete, and the increase in
andcontent
gel the increase in gel content
by the pozzolanic by the
reaction pozzolanic
of FA caused the reaction
strengthofdifference
FA caused the strength
between FA dif
ference between FA concrete and pure cement
concrete and pure cement concrete to gradually become smaller. concrete to gradually become smaller.

Figure6.6.Relationship
Figure Relationshipbetween FA concrete
between compressive
FA concrete strength
compressive and FA and
strength content.
FA content.

The compressive strength of manufactured sand concrete with different FA contents at

28 days and 90 days under both steam and standard curing is shown in Figure 7. From the
results, it is evident that the compressive strengths of pure cement concrete at 28 and 90 days
under steam curing was reduced by 6.8 and 7.2%, compared to those under standard curing.
This indicates that the thermal effect of steam curing was unfavorable to the late strength,
 The compressive strength of manufactured sand concrete with different FA contents
at 28 days and 90 days under both steam and standard curing is shown in Figure 7. From
the results, it is evident that the compressive strengths of pure cement concrete at 28 and
90 days under steam curing was reduced by 6.8 and 7.2%, compared to those under stand-
Materials 2023, 16, 1326 ard curing. This indicates that the thermal effect of steam curing was unfavorable 8 of 16 to the
late strength, which was consistent with conclusions from previous studies [4,17]. These
phenomena are mainly attributed to the uneven distribution of hydration products and
which was consistent
an uncompact with conclusions
microstructure at highfrom previous studies
temperatures [4,17]. These
[13,15,16]. phenomena
Concrete supplemented
are mainly
with FA hadattributed
lower to the uneven distribution
compressive strengths at of 28
hydration products and
days compared an uncompact
to those of pure cement
microstructure
concrete at high steam
(F0) under temperatures
curing.[13,15,16].
However, Concrete supplemented
at 90 days, with FA had
the compressive lower of F10
strength
compressive strengths at 28 days compared to those of pure cement concrete (F0) under
was close to F0, that of F20 was higher than F0, while that of F30 was still much lower than
steam curing. However, at 90 days, the compressive strength of F10 was close to F0, that of
that of F0. Notably, adding an appropriate amount of FA could improve the late strength
F20 was higher than F0, while that of F30 was still much lower than that of F0. Notably,
of concrete
adding under steam
an appropriate curing.
amount of We attribute
FA could this phenomenon
improve the late strengthto several possible
of concrete underreasons
(1)
steam curing. We attribute this phenomenon to several possible reasons: (1) Crystalline Pozzolanic
Crystalline FA filled the pores and reduced the content of large pores. (2) FA
reaction products
filled the pores were cross-bonded
and reduced the content ofwith
large the cement
pores. hydration
(2) Pozzolanic products,
reaction productswhich im-
proved
were the strength
cross-bonded withofthethe gel system.
cement hydration (3)products,
Steam curing advancedthe
which improved the occurrence
strength of the time of
gel system. (3) Steam curing advanced the occurrence time of the pozzolanic
the pozzolanic reaction of FA, increased the degree of pozzolanic reaction, produced reaction of FA, more
increased the degree of pozzolanic reaction, produced more calcium
calcium silicate hydrate and calcium aluminate hydrate, filled a large number of poressilicate hydrate and
calcium aluminate hydrate, filled a large number of pores and microcracks and improved
and microcracks and improved the compactness of the microstructure. Excessive FA sig-
the compactness of the microstructure. Excessive FA significantly reduced the late strength
nificantly reduced the late strength of the concrete. This was due to the large amount of
of the concrete. This was due to the large amount of FA, which reduced the true cement
FA, which
content, reduced
resulting in a the true cement
reduction content,
in hydration resulting
products. in a reduction
Additionally, in hydration
since most FA were prod-
ucts. Additionally, since most FA were crystalline phases, this made
crystalline phases, this made FA less reactive than cement [20,50–52], resulting in a reducedFA less reactive than
cement [20,50–52], resulting in a reduced gel content.
gel content.

Figure 7.7.Compressive
Figure Compressivestrength of concrete
strength withwith
of concrete different FA contents
different at 28 days
FA contents anddays
at 28 90 days.
and 90 days.
3.2. Microstructure
3.2. Microstructure
SEM images of manufactured sand concrete under the constant temperature stage of
steamSEM
curingimages
and theof late
manufactured sand concrete
stage are presented in Figure under the constant
8. To confirm temperature
that the pozzolanic stage of
steam
reactioncuring and theatlate
of FA occurred thestage aretemperature
constant presented stage,
in Figure 8. To confirm
we observed surface that the pozzolanic
morphology
reaction of FAatoccurred
of FA particles at the
the constant constant temperature
temperature stage. Resultsstage,
showed wethat
observed surface
the external mor-morphol-
phology
ogy of FA particles at the constant temperature stage. Results showed that theaexternal
of FA spherical particles of F20 concrete first changed from a smooth spherical to
speckled surface,
morphology before
of FA curing at
spherical a constant
particles temperature
of F20 concretefor 12 h
first (Figure 8a,b).
changed from Cement‘s
a smooth spher-
hydration products were precipitated on the surface of FA, indicating that FA exerted a
ical to a speckled surface, before curing at a constant temperature for 12 h (Figure 8a,b)
seeding effect. After 24 h of constant temperature curing, the surface of FA particles trans-
Cement‘s hydration products were precipitated on the surface of FA, indicating that FA
formed into a fibrous mesh containing a large number of pores (Figure 8c,d), indicating
exerted a seeding effect.
that the pozzolanic reactionAfter
of FA24hadh of constant
occurred. temperature
Pozzolanic curing,
reactions theunder
of FA surface of FA par-
steam
ticles
curing appeared at about 24 h of constant temperature curing, while the pozzolanic reaction 8c,d)
transformed into a fibrous mesh containing a large number of pores (Figure
indicating
appeared at that
aboutthe pozzolanic
7 days ◦ C curingof
under 20reaction FA had Our
[26,53,54]. occurred. Pozzolanic
results further reactions of FA
demonstrated
under steam
that steam curing
curing couldappeared at about
significantly 24 hthe
advance of pozzolanic
constant temperature curing, while the poz-
reaction of FA.
Occurrence of the pozzolanic reaction of FA under the constant temperature stage
during steam curing can be attributed to three reasons. Firstly, accelerated hydration of
cement under high-temperature curing and early production of large amounts of calcium
hydroxide provides the preconditions for pozzolanic reactions. Secondly, a large amount of
 zolanic reaction appeared at about 7 days under 20 °C curing [26,53,54]. Our results fur-
ther demonstrated that steam curing could significantly advance the pozzolanic reaction
of FA.
Occurrence of the pozzolanic reaction of FA under the constant temperature stage
Materials 2023, 16, 1326 during steam curing can be attributed to three reasons. Firstly, accelerated hydration 9 ofof
16

cement under high-temperature curing and early production of large amounts of calcium
hydroxide provides the preconditions for pozzolanic reactions. Secondly, a large amount
calcium
of calcium hydroxide
hydroxideproduced
producedover overaashortshortperiod
periodof oftime
time increased
increased alkalinity of of the
thepore
pore
solution. The
solution. Thehigh
highpH pHfacilitated
facilitated the
the dissolution
dissolution of of active
active SiO
SiO22 and
andAl Al22OO33ininthe
theglassy
glassyFA, FA,
thereby accelerating
thereby accelerating pozzolanic
pozzolanic reactions
reactions [51].
[51].Lastly,
Lastly, the
therapidly
rapidlyproduced
producedhydration
hydrationand and
pozzolanicreaction
pozzolanic reactionproducts
productswereweredistributed
distributedin inaafibrous
fibrousnetwork
networkon onthethesurface
surfaceof ofFAFA
(Figure 8c,d), subsequently forming a large number of pores, which facilitated entry of Ca 2+
(Figure subsequently forming a large number of pores, which facilitated entry of
and −
OH OH into thethe
interior of FA at aatlater stage andand
promoted thethe
pozzolanic reaction.
Ca 2+ and − into interior of FA a later stage promoted pozzolanic reaction.
Under steam curing,
Under curing,pure
purecement
cementconcrete
concrete exhibited
exhibited coarse andand
coarse numerous
numerous microscopic
micro-
pores, pores,
scopic whichwhichwere accompanied
were accompanied by uneven distribution
by uneven of hydration
distribution of hydrationproducts
productsandandthe
presence
the of microcracks
presence of microcracks(Figure 8e,f).8e,f).
(Figure These may may
These explain the loss
explain theofloss
lateofstrength in steam-
late strength in
cured concrete [6,11,15]. In FA concrete, the inert FA mainly filled
steam-cured concrete [6,11,15]. In FA concrete, the inert FA mainly filled the pores andthe pores and reduced the
large pores
reduced thecontent. The gel
large pores generated
content. The by gelthe pozzolanic
generated by reaction of glassyreaction
the pozzolanic FA not only filled
of glassy
the pores, but also cross-bonded with the C-S-H (Figure 8d), C-(A)-S-H
FA not only filled the pores, but also cross-bonded with the C-S-H (Figure 8d), C-(A)-S-H (Figure 8h), needle-
like AFt8h),
(Figure (Figure 8g) and
needle-like AFtplate-like
(Figure 8g)CHand (Figure 8d). Overall,
plate-like CH (Figure these phenomena
8d). Overall, thesenot phe-
only
increased
nomena notthe strength
only of the
increased thegelstrength
system,ofbut thealso
gel improved
system, but microstructural
also improvedcompactness.
microstruc-
That decreased the negative impact of steam curing on the microstructure,
tural compactness. That decreased the negative impact of steam curing on the microstruc- which in turn
reduced the thermal damage to the strength.
ture, which in turn reduced the thermal damage to the strength.

Figure 8. Cont.
Materials2023,
Materials 16,x1326
2023,16, FOR PEER REVIEW 1010ofof16
16

Figure 8. SEM image of concrete. Images (a–d) are F20 maintained under steam curing for 6 h, 12 h,
Figure 8. SEM image of concrete. Images (a–d) are F20 maintained under steam curing for 6 h, 12 h,
24 h and 36 h, respectively. Images (e,f) are 24 h and 28 days after steam curing of F0. Images (g,h)
24 h and 36 h, respectively. Images (e,f) are 24 h and 28 days after steam curing of F0. Images (g,h) are
are 28 days and 90 days after steam curing of F20.
28 days and 90 days after steam curing of F20.
3.3.
3.3.XRD
XRDAnalyses
Analyses
Next,
Next, weweused
usedXRD XRDto toexplore
explorethethe effect
effect of
of FAFA on
on the
the hydration
hydration of of manufactured
manufactured
sand concrete during the constant temperature stage of steam
sand concrete during the constant temperature stage of steam curing. XRD curing. XRD patterns of F20
patterns of
concrete cured at constant temperature for 6, 12, 24 and 36 h are shown
F20 concrete cured at constant temperature for 6, 12, 24 and 36 h are shown in Figure in Figure 9. At 69.
hAtof6constant temperature
h of constant temperaturecuring, we observed
curing, we observedboth both
unhydrated C3S (PDF
unhydrated #49-0442)
C3 S (PDF and
#49-0442)
Cand
2S (PDF #49-1673)
C2 S (PDF in allinfour
#49-1673) groups
all four of concrete
groups of concretespecimens. The The
specimens. diffraction
diffraction peak of
peak
ettringite (PDF
of ettringite #41-1451)
(PDF can also
#41-1451) can be observed,
also whichwhich
be observed, was a was
hydration productproduct
a hydration of trical-
of
cium aluminate (C A). Gismondine (PDF #39-1373) can be observed, but
tricalcium aluminate (C3 A). Gismondine (PDF #39-1373) can be observed, but the intensity
3 the intensity of
its
of diffraction peak
its diffraction peakwas low,
was which
low, which waswasa product
a productofofcement
cementhydration
hydration[55].[55].Moreover,
Moreover,
CH
CH(PDF(PDF#04-0733)
#04-0733)diffraction
diffractionpeaks
peakswere
werepresent.
present.The Theintensity
intensityof
ofthe
thecharacteristic
characteristicpeak peak
of
ofCH
CHin inF30
F30was
was the
the lowest,
lowest, andand the
the intensity
intensity ofof the
the characteristic
characteristic peak
peak ofof CC33SSin
inF0
F0was
was
the
the lowest.
lowest. According to to the
theprinciple
principleofofthethehydration
hydration reaction
reaction of Portland
of Portland cement,
cement, ce-
cement
ment hydration
hydration degree
degree can becan be expressed
expressed by theby the content
content of CH [50,53].
of CH [50,53]. The cement
The actual actual cement
content
content in FA concrete
in FA concrete was reduced,
was reduced, resulting resulting in less
in less early early CH production
CH production and thereforeandlower
therefore
early
compressive
lower strength. strength.
early compressive
The Intensity of the diffraction peaks of C3S and C2S significantly decreased in all
four groups of specimens, while that of the diffraction peaks of CH increased after 12 h of
constant temperature curing. This indicated the intense occurrence of the hydration reac-
tion of the cement. The intensity of the CH diffraction peak in F30 was not significantly
 CH diffraction peak intensity in FA concrete may be attributed to the following reasons:
Firstly, addition of FA resulted in lower cement content in concrete and reduced the
amount of CH produced by hydration. Secondly, part of the CH produced by cement hy-
dration started to be consumed by the pozzolanic reaction of FA, as evidenced by SEM
images. We found no obvious change in the intensity of diffraction peaks of CH, C 2S and
Materials 2023, 16, 1326 C3S across all four groups of samples after 36 h under constant temperature curing.11This
of 16
indicated that the hydration reaction became significantly slower.

Figure
Figure 9. XRD
XRD atatconstant
constanttemperature
temperature under
under steam
steam curing.
curing. E–Ettringite;
E–Ettringite; CH–Portlandite;
CH–Portlandite; Q–
Q–Quartz;
Quartz; C–Calcite; CS–C S/C S; G–Gismondine.
C–Calcite; CS–C3 S/C2 S; G–Gismondine.
3 2

The Intensity
Profiles of XRDofpatterns
the diffraction peaks
during later of C3of
stages S and C2 S significantly
steam-cured decreased
manufactured sand in all
con-
crete are illustrated in Figure 10. From the results, it was observed that the intensity of theh
four groups of specimens, while that of the diffraction peaks of CH increased after 12
of constantpeak
diffraction temperature curing. with
of CH decreased This an
indicated
increasethe intense
in FA occurrence
content. Notably,ofhigher
the hydration
FA con-
reaction
tent of the
resulted in cement.
more CHThe intensity of athe
consumption, CH diffraction
higher peak
degree of FA in F30 was
reaction, notissignificantly
which consistent
reduced,
with indicating under
the regularity that the pozzolanic
ambient reaction of
temperature FA did
curing notThe
[51]. appear. The diffraction
diffraction peaks ofpeaks
C 2S
of CC3 S
and 3Sand
wereCstill
2 S were still observed
observed at 28 daysatand
a constant
90 days,temperature of 24ofh,the
but the intensity butdiffraction
the intensitypeakof
their diffraction peaks was significantly lower. The intensity of the diffraction
at 90 days was lower. This indicated that C2S and C3S were still slowly hydrated, and the peak of CH
continued
CH produced to increase, with the largest
by late hydration increasethe
also provided observed in F0.
conditions forThe
thelow increase of
persistence in the
CH
diffraction peak
pozzolanic reaction. intensity in FA concrete may be attributed to the following reasons: Firstly,
addition of FA resulted in lower cement content in concrete and reduced the amount of CH
produced by hydration. Secondly, part of the CH produced by cement hydration started to
be consumed by the pozzolanic reaction of FA, as evidenced by SEM images. We found
no obvious change in the intensity of diffraction peaks of CH, C2 S and C3 S across all four
groups of samples after 36 h under constant temperature curing. This indicated that the
hydration reaction became significantly slower.
Profiles of XRD patterns during later stages of steam-cured manufactured sand con-
crete are illustrated in Figure 10. From the results, it was observed that the intensity of the
diffraction peak of CH decreased with an increase in FA content. Notably, higher FA content
resulted in more CH consumption, a higher degree of FA reaction, which is consistent with
the regularity under ambient temperature curing [51]. The diffraction peaks of C2 S and
C3 S were still observed at 28 days and 90 days, but the intensity of the diffraction peak
at 90 days was lower. This indicated that C2 S and C3 S were still slowly hydrated, and
the CH produced by late hydration also provided the conditions for the persistence of the
pozzolanic reaction.
 Materials
Materials 2023,
2023, 16,16,
x xFOR
FORPEER
PEERREVIEW
REVIEW 12 of
1216of 16
Materials 2023, 16, 1326 12 of 16

Figure 10. XRD after steam curing. E–Ettringite; CH–Portlandite; Q–Quartz; C–Calcite; CS–C3S/C2S;
Figure10.
Figure XRDafter
10.XRD
G–Gismondine. after steam
steam curing.
curing. E–Ettringite;
E–Ettringite;CH–Portlandite;
CH–Portlandite;Q–Quartz; C–Calcite;
Q–Quartz; CS–C
C–Calcite; 3 S/C
CS–C 2 S; 2S;
3S/C
G–Gismondine.
G–Gismondine.
3.4. T2 spectra
3.4. T2 spectra
3.4. TThe2 spectra
area integral of the T2 spectral distribution curve reflected the pore volume of the
The area integral of the T2 spectral distribution curve reflected the pore volume of the
The
concrete, area
concrete, and and integral
the of thetime
the relaxation
relaxation T2 spectral
time distribution
represented
represented thethe pore
pore curve
sizesize reflected
[45,48].
[45,48]. WeWe thedivided
porethe
divided volume
the
poreporeof the
size
size to
concrete, gel pores
and the (d < 10
relaxation nm), capillary
time pores
represented (10 nm–1000
the pore nm)
size
to gel pores (d < 10 nm), capillary pores (10 nm–1000 nm) and large pores (d > 1000 nm) [49]. and
[45,48]. largeWe pores (d
divided > 1000
the pore
nm)
size [49]. Distribution
to gel pores
Distribution curves curves
(d <of10Tnm), of T spectra of steam-cured manufactured
capillary pores (10 nm–1000 nm) and large pores (d > 1000
2 sand concrete at
2 spectra of steam-cured manufactured sand concrete at 28 and
28
90 and
nm) [49].
days 90with
daysdifferent
with different
Distribution curves FA
of Tcontents
FA contents 2 spectra are
are shown ofshown
steam-cured
in Figurein Figure 11. Based
11.manufactured
Based on the onresults,
the results,
sand concrete
it wasit at
was
28 andobserved
observed thatthat
90 days with
the the first
different
first peakpeakinFAin theT2Tspectrum
thecontents 2 spectrum
are shown ininFAFAin concrete
Figure had
concrete had significantly
11. Based on the higher
significantly results, it
higher
relaxation-signal
was intensity
observed thatintensity
relaxation-signal the first andand spectral
peakspectral area
in the Tarea than
2 spectrum
those
than those of
in FA pure cement
concrete
of pure cement concrete.
hadconcrete. The
Thepeak
significantly higher
peak
signals of the second
signals of the second
relaxation-signal and third
and third
intensity peaks shifted
peaks shifted
and spectral to
area than the
to the left. This
left.ofThis
those indicated
pure indicated that the
cement concrete. volume
Theof
that the volume peak
both
of both
signals gelofgel
and
thecapillary
and capillary
second pores
and increased,
pores
third increased, while
peaks shifted the
while pore
to the
the poresizeThis
left. of large
size of pores
large
indicated decreased
pores in FA
thatdecreased
the volume in of
concrete
FA concrete compared to pure cement concrete. Based on SEM images, we
weattributed this
both gel and compared
capillary pores to pure cement
increased, concrete.
while the Based
poreon SEM
size images,
of large pores attributed
decreasedthis in FA
phenomenon
phenomenon to
tothe
thepozzolanic
pozzolanic reaction
reaction and
andfilling
filling effect
effect of FA. Increase in
inage caused
causedaathis
concrete compared to pure cement concrete. Based on of SEM FA.images,
Increasewe age
attributed
decrease
decrease in in the peak signal
thetopeak signalintensity
intensityof ofTT22spectra
spectraininall allfour
four groupsofof specimens. Moreo-
phenomenon the pozzolanic reaction and filling effectgroups
of FA. Increasespecimens. Moreover,
in age caused a
ver, the peak signals of the second and third peaks moved
the peak signals of the second and third peaks moved to the left, with a large left shiftto the left, with a large left shift
in
decrease in the peak signal intensity of T2 spectra in all four groups of specimens. Moreo-
in the FA concrete. These results indicated that both the subsequent
the FA concrete. These results indicated that both the subsequent continuous hydration continuous hydration
ver,
and the peak signals of the second andthethird peaks moved to the left, withsize.
a large left
con-shift
and pozzolanic reaction
pozzolanic reaction could
could reduce
reduce the porepore content
content andand refine
refine thetheporepore
size. FA FAconcrete
in the FA concrete.
crete These results indicated that bothreaction,
the subsequent continuous hydration
had ahad dual a dual
effecteffect of hydration
of hydration and pozzolanic
and pozzolanic reaction, which which subsequently
subsequently gener-a
generated
and
ated pozzolanic
a significant
significant reaction
variationvariation could reduce
in thesignal
in the peak the
peak signal pore
intensity content
intensity and refine
and location.
and location. the pore size. FA con-
crete had a dual effect of hydration and pozzolanic reaction, which subsequently gener-
ated a significant variation in the peak signal intensity and location.

Figure
Figure 11.
11.TT2 2spectrum
spectrumdistribution
distributioncurves ofof
curves concrete with
concrete different
with FAFA
different contents. (a) (a)
contents. At At
28 days,
28 days,
(b)
(b) At
At 90
90 days.
days.

FigureThe
11. T
The 2 spectrum
area integraldistribution
integral of curves of
theTT2 2spectrum
ofthe spectrum concrete with
distribution different
curves
distribution curvesof of FA contents.sand
manufactured
manufactured (a)sand
Atconcrete
28con-
days,
(b) At
with 90 days.
different FA contents under steam curing are shown in Figure 12. Summarily,
crete with different FA contents under steam curing are shown in Figure 12. Summarily, FA con-
crete
FA had higher
concrete total pore
had higher totalvolumes under under
pore volumes steam steam
curingcuring
than pure
thancement concrete,
pure cement which
concrete,
wasThe
which area
was integral
consistent
consistent of the
with T2 spectrum
with conclusions
conclusions fromdistribution
from previous studies
previous curves of ◦manufactured
underunder
studies 20 C20
curing sand
[27,56].
°C curing con-
This
[27,56].
crete with different FA contents under steam curing are shown in Figure 12. Summarily,
FA concrete had higher total pore volumes under steam curing than pure cement concrete,
which was consistent with conclusions from previous studies under 20 °C curing [27,56].
 This phenomenon was attributed to the increase in the real water-to-binder rati
concrete. Total pore volume first decreased, then increased when the FA content
Materials 2023, 16, x FOR PEER REVIEW 13 of 16
crete increased from 10 to 30%, indicating that there was an optimum value of F
Materials 2023, 16, 1326 13 of 16
tion. Combined with SEM and XRD images, it was evident that FA‘s filling eff
pozzolanic effect caused a decrease in large pore content and an increase in mic
This phenomenon was attributed to the increase in the real water-to-binder ratio in FA
tural compactness, thereby favorably affecting the strength. Increase in age ca
phenomenon
concrete. waspore
Total attributed
volumeto the increase
first in the then
decreased, real water-to-binder
increased when ratio in FA
the FA concrete.
content in con-
marked
Total pore reduction
volume first in total pore
decreased, then volume
increased of
whenthetheFAFAconcrete
content inrelative
concrete to that of pure
increased
crete increased from 10 to 30%, indicating that there was an optimum value of FA addi-
concrete,
from
tion. 10 indicating
to 30%,
Combined indicating
with SEM
that the
that pozzolanic
there
and XRD wasimages, reaction
an optimumit was
during
value of FAlater
evident
stages
addition.
that
significantly
FA‘sCombined
filling effect and
with SEM and
the reduction XRD images,
in caused
the pore itvolume.in large pore content and an increase ineffect
was evident that FA‘s filling effect and pozzolanic
pozzolanic effect a decrease microstruc-
caused a decrease in large pore content and an increase in microstructural compactness,
tural
therebycompactness, thereby
favorably affecting thefavorably affecting
strength. Increase in the strength.
age caused Increase
a marked in age in
reduction caused a
marked
total porereduction
volume ofin total
the pore volume
FA concrete of to
relative the FAofconcrete
that relative
pure cement to that
concrete, of pure cement
indicating
concrete,
that the pozzolanic reaction during later stages significantly affected the reduction in theaffected
indicating that the pozzolanic reaction during later stages significantly
the
porereduction
volume. in the pore volume.

Figure 12. Integral area of the T2 spectrum.

The ratio of the area of the first, second and third peaks to the total integrated
Figure 12.
the T212.
Figure Integralarea
spectra
Integral ofarea
of of
thethe
T2 T
manufactured 2 spectrum.
spectrum. sand concrete with different FA contents is illust
Figure 13. Summarily, the largest percentage of the first peak area indicated that
The
Theratio
ratioofofthe
thearea of of
area thethe
first, second
first, and and
second thirdthird
peakspeaks
to the to
total
theintegrated area in area in
total integrated
theTpores
the spectrainofthe steam-cured
manufactured sand manufactured
concrete with sand
different concrete
FA contents were
is both
illustratedgel
in and in c
the T22 spectra of manufactured sand concrete with different FA contents is illustrated
pores.13.
Figure Notably,
Summarily, an the
increase
largestin FA content
percentage of thegenerated a corresponding
first peak area indicated that most increase
of in
Figure 13. Summarily, the largest percentage of the first peak area indicated that most of
the pores
portion in the steam-cured manufactured sand concrete were both gel and capillary pores.
the poresofinthethefirst peak area,
steam-cured and a decrease
manufactured sand in the proportion
concrete were both of
Notably, an increase in FA content generated a corresponding increase in the proportion
gelthe
andsecond an
capillary
peak areas.
pores. Notably,These results in
an increase demonstrated
FA incontent that theavolume
generated proportion of gel pores
of the first peak area, and a decrease the proportion of thecorresponding
second and third increase in the
peak areas. pro-
pillary
portion pores
of the in manufactured
first peak area, and sand
a concrete
decrease in increased
the while
proportion the
of
These results demonstrated that the volume proportion of gel pores and capillary pores inthe proportion
second and of larg
third
decreased.
peak areas. These
manufactured results demonstrated
sand concrete increased whilethat the volume
the proportion proportion
of large of gel pores and ca-
pores decreased.
pillary pores in manufactured sand concrete increased while the proportion of large pores
decreased.

Figure The
Figure13.13. proportion
The of each
proportion ofpeak.
each peak.
Figure 13. The proportion of each peak.
4. Conclusions
4. Conclusions
Materials 2023, 16, 1326 14 of 16

4. Conclusions
In summary, our findings indicate that the addition of appropriate amounts of FA can
alleviate the negative effects of steam curing on the long-term strength of manufactured
sand concrete. This is expected to facilitate production of precast concrete beams in cold
regions. The following main conclusions are drawn from this study:
(1) The steam curing time required to reach the demoulding and prestress tensioning
strengths of F30 was 100% and 50% longer than that of F0, respectively. This phe-
nomenon was caused by the dilution effect of FA.
(2) The intensity of the diffraction peak of CH in FA concrete decreased after 24 h of
constant temperature under steam curing. FA particles were dissolved and the smooth
spherical surface changed to a fibrous network, which proved that the pozzolanic
reaction of FA occurred, indicating that steaming could remarkably advance the
pozzolanic reaction time of FA.
(3) Appropriate amounts of FA need to be added in the steam-cured manufactured sand
concrete. The reason was that the filling effect and pozzolanic reaction of FA would
result in lower large pore content, increased microstructural compactness and higher
strength of the gel system, which in turn reduced the adverse effect of steam curing
on the microstructure, and therefore had a beneficial effect on the long-term strength.
The amount of FA admixture in steam-cured manufactured sand concrete was suitable
to be controlled at 20%.
Although it was proved that the pozzolanic reaction of FA could happen at 24 h of
steam curing, the following aspects still need further study: (1) Reaction kinetics of FA
concrete under steam curing [19]. (2) Effect of FA on other properties of steam-cured
manufactured sand concrete, such as chloride ion impermeability and frost resistance.

Author Contributions: Methodology, Y.D.; investigation, X.W.; data curation, Z.L.; writing—original
draft preparation, Y.D.; writing—review and editing, Z.L.; visualization, X.W.; supervision, Q.W.;
funding acquisition, Q.W. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the National Natural Science Foundation of China (No.
51768033), Excellent Youth Foundation of Gansu Province (21JR7RA344), Transportation Science and
Technology Project of Gansu (19-K275).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Liu, Y.; Xie, S.; Zhou, J.; Li, J. Some factors affecting early compressive strength of steam-curing concrete with ultrafine fly ash.
Cem. Concr. Res. 2001, 31, 1455–1458. [CrossRef]
2. Shi, J.; Liu, B.; Zhou, F.; Shen, S.; Dai, J.; Ji, R.; Tan, J. Heat damage of concrete surfaces under steam curing and improvement
measures. Constr. Build. Mater. 2020, 252, 119104. [CrossRef]
3. Gonzalez-Corominas, A.; Etxeberria, M.; Poon, C.S. Influence of steam curing on the pore structures and mechanical properties of
fly-ash high performance concrete prepared with recycled aggregates. Cem. Concr. Compos. 2016, 71, 77–84. [CrossRef]
4. Ramezanianpour, A.A.; Khazali, M.H.; Vosoughi, P. Effect of steam curing cycles on strength and durability of SCC: A case study
in precast concrete. Constr. Build. Mater. 2013, 49, 807–813. [CrossRef]
5. Türkel, S.; Alabas, V. The effect of excessive steam curing on Portland composite cement concrete. Cem. Concr. Res. 2005, 35,
405–411. [CrossRef]
6. Shi, J.; Liu, B.; Zhou, F.; Shen, S.; Guo, A.; Xie, Y. Effect of steam curing regimes on temperature and humidity gradient,
permeability and microstructure of concrete. Constr. Build. Mater. 2021, 281, 122562. [CrossRef]
7. Ba, M.; Qian, C.; Guo, X.; Han, X. Effects of steam curing on strength and porous structure of concrete with low water/binder
ratio. Constr. Build. Mater. 2011, 25, 123–128. [CrossRef]
8. Taylor, H.F.W.; Famy, C.; Scrivener, K.L. Delayed ettringite formation. Cem. Concr. Res. 2001, 31, 683–693. [CrossRef]
Materials 2023, 16, 1326 15 of 16

9. Glasser, F.; Damidot, D.; Atkins, M. Phase development in cement in relation to the secondary ettringite problem. Adv. Cem. Res.
1995, 7, 57–68. [CrossRef]
10. Yan, P.; Qin, X. The effect of expansive agent and possibility of delayed ettringite formation in shrinkage-compensating massive
concrete. Cem. Concr. Res. 2001, 31, 335–337. [CrossRef]
11. Shi, J.; Liu, B.; Wu, X.; Qin, J.; Jiang, J.; He, Z. Evolution of mechanical properties and permeability of concrete during steam
curing process. J. Build. Eng. 2020, 32, 101796. [CrossRef]
12. GB/T 37439-2019; Precast Post-Tensioned Prestressed Concrete Simply-Supported Beam of High-Speed Railway. Chinese Standard
Press: Beijing, China, 2019. (In Chinese)
13. Gallucci, E.; Zhang, X.; Scrivener, K.L. Effect of temperature on the microstructure of calcium silicate hydrate (C-S-H). Cem. Concr.
Res. 2013, 53, 185–195. [CrossRef]
14. Lothenbach, B.; Winnefeld, F.; Alder, C.; Wieland, E.; Lunk, P. Effect of temperature on the pore solution, microstructure and
hydration products of Portland cement pastes. Cem. Concr. Res. 2007, 37, 483–491. [CrossRef]
15. Zou, C.; Long, G.; Xie, Y.; He, J.; Ma, C.; Zeng, X. Evolution of multi-scale pore structure of concrete during steam-curing process.
Micropor. Mesopor. Mat. 2019, 288, 109566. [CrossRef]
16. Cao, Y.; Detwiler, R.J. Backscattered electron imaging of cement pastes cured at elevated temperatures. Cem. Concr. Res. 1995, 25,
627–638. [CrossRef]
17. Shi, J.; Liu, B.; Shen, S.; Tan, J.; Dai, J.; Ji, R. Effect of curing regime on long-term mechanical strength and transport properties of
steam-cured concrete. Constr. Build. Mater. 2020, 255, 119407. [CrossRef]
18. Khatib, J.; Jahami, A.; Kordi, A.; Sonebi, M.; Malek, Z.; Elchamaa, R.; Dakkour, S. Effect of municipal solid waste incineration
bottom ash (MSWI-BA) on the structural performance of reinforced concrete (RC) beams. J. Eng. Des. Technol. 2021. [CrossRef]
19. Hemalatha, T.; Ramaswamy, A. A review on fly ash characteristics—Towards promoting high volume utilization in developing
sustainable concrete. J. Clean. Prod. 2017, 147, 546–559. [CrossRef]
20. Li, Z.; Xu, G.; Shi, X. Reactivity of coal fly ash used in cementitious binder systems: A state-of-the-art overview. Fuel 2021,
301, 121031. [CrossRef]
21. Liu, C.; Zhang, M. Effect of curing temperature on hydration, microstructure and ionic diffusivity of fly ash blended cement
paste: A modelling study. Constr. Build. Mater. 2021, 297, 123834. [CrossRef]
22. Kim, J.H.; Noemi, N.; Shah, S.P. Effect of powder materials on the rheology and formwork pressure of self-consolidating concrete.
Cem. Concr. Compos. 2012, 34, 746–753. [CrossRef]
23. Chindaprasirt, P.; Homwuttiwong, S.; Sirivivatnanon, V. Influence of fly ash fineness on strength, drying shrinkage and sulfate
resistance of blended cement mortar. Cem. Concr. Res. 2004, 34, 1087–1092. [CrossRef]
24. Kwan, A.K.H.; Chen, J.J. Adding fly ash microsphere to improve packing density, flowability and strength of cement paste.
Powder. Technol. 2013, 234, 19–25. [CrossRef]
25. He, J.; Long, G.; Ma, K.; Xie, Y. Influence of fly ash or slag on nucleation and growth of early hydration of cement. Thermochim.
Acta. 2021, 701, 178964. [CrossRef]
26. Park, B.; Choi, Y.C. Hydration and pore-structure characteristics of high-volume fly ash cement pastes. Constr. Build. Mater. 2021,
278, 122390. [CrossRef]
27. Chindaprasirt, P.; Jaturapitakkul, C.; Sinsiri, T. Effect of fly ash fineness on compressive strength and pore size of blended cement
paste. Cem. Concr. Compos. 2005, 27, 425–428. [CrossRef]
28. Cao, C.; Sun, W.; Qin, H. The analysis on strength and fly ash effect of roller-compacted concrete with high volume fly ash. Cem.
Concr. Res. 2000, 30, 71–75. [CrossRef]
29. Mei, J.; Tan, H.; Li, H.; Ma, B.; Liu, X.; Jiang, W.; Zhang, T.; Li, X. Effect of sodium sulfate and nano-SiO2 on hydration and
microstructure of cementitious materials containing high volume fly ash under steam curing. Constr. Build. Mater. 2018, 163,
812–825. [CrossRef]
30. Liu, M.; Tan, H.; He, X. Effects of nano-SiO2 on early strength and microstructure of steam-cured high volume fly ash cement
system. Constr. Build. Mater. 2019, 194, 350–359. [CrossRef]
31. Alghazali, H.H.; Aljazaeri, Z.R.; Myers, J.J. Effect of accelerated curing regimes on high volume Fly ash mixtures in precast
manufacturing plants. Cem. Concr. Res. 2020, 131, 105913. [CrossRef]
32. Liu, B.; Xie, Y.; Li, J. Influence of steam curing on the compressive strength of concrete containing supplementary cementing
materials. Cem. Concr. Res. 2005, 35, 994–998. [CrossRef]
33. Hanehara, S.; Tomosawa, F.; Kobayakawa, M.; Hwang, K. Effects of water/powder ratio, mixing ratio of fly ash, and curing
temperature on pozzolanic reaction of fly ash in cement paste. Cem. Concr. Res. 2001, 31, 31–39. [CrossRef]
34. Yang, J.; Hu, H.; He, X.; Su, Y.; Wang, Y.; Tan, H.; Pan, H. Effect of steam curing on compressive strength and microstructure of
high volume ultrafine fly ash cement mortar. Constr. Build. Mater. 2021, 266, 120894. [CrossRef]
35. Yazıcı, H.; Aydın, S.; Yiğiter, H.; Baradan, B. Effect of steam curing on class C high-volume fly ash concrete mixtures. Cem. Concr.
Res. 2005, 35, 1122–1127. [CrossRef]
36. Jiang, P.; Jiang, L.; Zha, J.; Song, Z. Influence of temperature history on chloride diffusion in high volume fly ash concrete. Constr.
Build. Mater. 2017, 144, 677–685. [CrossRef]
37. Zhang, G.; Wei, Q.; Ding, Q.; Wang, A.; Liu, K. Effect of curing temperature and fly ash content on the hydration and microstructure
of fly ash–cement pastes. J. Sustain. Cem-Based 2018, 7, 372–383. [CrossRef]
Materials 2023, 16, 1326 16 of 16

38. Chindaprasirt, P.; Chotithanorm, C.; Cao, H.T.; Sirivivatnanon, V. Influence of fly ash fineness on the chloride penetration of
concrete. Constr. Build. Mater. 2007, 21, 356–361. [CrossRef]
39. GB 175-2007; Common Portland Cement. Chinese Standard Press: Beijing, China, 2007. (In Chinese)
40. GB/T 14684-2011; Sand for Construction. Chinese Standard Press: Beijing, China, 2011. (In Chinese)
41. Duan, Y.; Wang, Q.; Yang, Z.; Cui, X.; Liu, F.; Chen, H. Research on the effect of steam curing temperature and duration on the
strength of manufactured sand concrete and strength estimation model considering thermal damage. Constr. Build. Mater. 2022,
315, 125531. [CrossRef]
42. GB/T 50081-2019; Standard for Test Methods of Concrete Physical and Mechanical Properties. Chinese Standard Press: Beijing,
China, 2019. (In Chinese)
43. Liu, K.; Wang, S.; Quan, X.; Duan, W.; Nan, Z.; Wei, T.; Xu, F.; Li, B. Study on the mechanical properties and microstructure of
fiber reinforced metakaolin-based recycled aggregate concrete. Constr. Build. Mater. 2021, 294, 123554. [CrossRef]
44. Rifai, H.; Staude, A.; Meinel, D.; Illerhaus, B.; Bruno, G. In-situ pore size investigations of loaded porous concrete with
nondestructive methods. Cem. Concr. Res. 2018, 111, 72–80. [CrossRef]
45. Stelzner, L.; Powierza, B.; Oesch, T.; Dlugosch, R.; Weise, F. Thermally-induced moisture transport in high-performance concrete
studied by X-ray-CT and 1H-NMR. Constr. Build. Mater. 2019, 224, 600–609. [CrossRef]
46. Edelmann, M.; Zibold, T.; Grunewald, J. Generalised NMR-moisture correlation function of building materials based on a
capillary bundle model. Cem. Concr. Res. 2019, 119, 126–131. [CrossRef]
47. Valori, A.; McDonald, P.J.; Scrivener, K.L. The morphology of C-S-H: Lessons from 1H nuclear magnetic resonance relaxometry.
Cem. Concr. Res. 2013, 49, 65–81. [CrossRef]
48. Shen, Y.; Wang, Y.; Wei, X.; Jia, H.; Yan, R. Investigation on meso-debonding process of the sandstone–concrete interface induced
by freeze–thaw cycles using NMR technology. Constr. Build. Mater. 2020, 252, 118962. [CrossRef]
49. Gao, F.; Tian, W.; Cheng, X. Investigation of moisture migration of MWCNTs concrete after different heating-cooling process by
LF-NMR. Constr. Build. Mater. 2021, 288, 123146. [CrossRef]
50. Papadakis, V.G. Effect of fly ash on Portland cement systems: Part I. Low-calcium fly ash. Cem. Concr. Res. 1999, 29, 1727–1736.
[CrossRef]
51. Zhang, Y.M.; Sun, W.; Yan, H.D. Hydration of high-volume fly ash cement pastes. Cem. Concr. Compos. 2000, 22, 445–452.
[CrossRef]
52. Durdziński, P.T.; Dunant, C.F.; Haha, M.B.; Scrivener, K.L. A new quantification method based on SEM-EDS to assess fly ash
composition and study the reaction of its individual components in hydrating cement paste. Cem. Concr. Res. 2015, 73, 111–122.
[CrossRef]
53. Sakai, E.; Miyahara, S.; Ohsawa, S.; Lee, S.-H.; Daimon, M. Hydration of fly ash cement. Cem. Concr. Res. 2005, 35, 1135–1140.
[CrossRef]
54. Baert, G.; Hoste, S.; De Schutter, G.; De Belie, N. Reactivity of fly ash in cement paste studied by means of thermogravimetry and
isothermal calorimetry. J. Therm. Anal. Calorim. 2008, 94, 485–492. [CrossRef]
55. Long, Z.; Zhang, R.; Wang, Q.; Xie, C.; Zhang, J.; Duan, Y. Mechanism analysis of strength evolution of concrete structure in saline
soil area based on 15-year service. Constr. Build. Mater. 2022, 332, 127281. [CrossRef]
56. Zhao, H.; Qin, X.; Liu, J.; Zhou, L.; Tian, Q.; Wang, P. Pore structure characterization of early-age cement pastes blended with
high-volume fly ash. Constr. Build. Mater. 2018, 189, 934–946. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.