materials

Article
Complex Effect of Concrete Composition on the
Thermo-Mechanical Behaviour of Mass Concrete
Barbara Klemczak 1, * , Maciej Batog 2 , Zbigniew Giergiczny 1 and Aneta Żmij 1
1 Faculty of Civil Engineering, Silesian University of Technology, Gliwice 44-100, Poland;
Zbigniew.Giergiczny@polsl.pl (Z.G.); Aneta.Zmij@polsl.pl (A.Z.)
2 Górażdże Cement S.A., Chorula 47-316, Poland
* Correspondence: Barbara.Klemczak@polsl.pl (B.K.)




Received: 14 September 2018; Accepted: 1 November 2018; Published: 7 November 2018


Abstract: The current work presents the complex investigation of the influence of cement and
aggregate type on the thermo-mechanical behavior of mass concrete. Six types of cement with
different amounts of non-clinker constituents and four types of aggregates are used in experimental
tests. Particular attention was given to the low clinker cements with high amounts of siliceous fly
ash and ground blast furnace slag. The experimental research covered the determination of thermal,
mechanical, and rheological properties of early age concrete with different constituents. Experimental
results have been used both to validate the numerical model and analysis of exemplary foundation
slab. The results confirm the importance of the concrete mix composition and it has been shown
that the early-age volume deformation and possible cracking is the result of the concerted action
of thermal and mechanical properties of concrete. The obtained results indicate granite as the best
aggregate for mass concrete. Considering the type of cement, much better behaviour of mass concrete
has been noted for cements with fly ash and composite cements containing both fly ash and slags
than cements only with slag.

Keywords: mass concrete; early age properties; low clinker cement; aggregate; hardening temperature;
mechanical properties; shrinkage; creep; early age stresses

1. Introduction
The temperature rise during cement hydration is still an open problem in the construction of mass
concrete structures. Due to the considerable thickness of such structures, the amount of heat generated
is significant and the core of the concrete member reaches particularly high temperatures, which is
also higher than the temperature of its surface layers. Such non-uniform temperature distribution
and its variation in time induces thermal strains. Additionally, shrinkage strains are also observed
in early age concrete because of the hydration process and moisture exchange with the environment.
Consequently, stresses arise in the concrete structure, which, if not properly controlled, can result in
cracking, affecting service life of the structure. This early age cracking of thermal and shrinkage origin
may initiate corrosion of reinforcement. Moreover, at a later age, the crack width may increase due to
environmental and mechanical loads.
The size of early age thermal-shrinkage volume changes and possible cracking depends both
on material and technological factors. Among others, concrete mix composition, type of concrete
constituents, concrete curing condition, the size of concrete volume, and structure restraint conditions
can be specified here [1,2]. In this respect, the concrete’s composition has a particularly significant
effect on the discussed volume changes [3–6]. Hence, one of the methods for reducing the negative
effects of self-heating of mass structures is the use of cements with moderate or low heat of hydration.
Thus, the most practical way to limit hydration heat and in consequence the hardening temperature

Materials 2018, 11, 2207; doi:10.3390/ma11112207 www.mdpi.com/journal/materials

Materials 2018, 11, 2207 2 of 18

is the replacement of Portland cement with some non-clinker constituents as siliceous fly ash (FA)
or ground blast furnace slag (GBFS). Such a replacement creates a binder which produces much less
heat [7,8]. There is no doubt that the drawback of cements with non-clinker constituents is the strength
gain, which is slower at early ages and may lessen the load-bearing capacity of the structure [9–11].
The slower strength gains at early ages is explained by the slower pozzolanic reaction of the slag
and fly ash compared to Portland clinker. However, it was proved that pozzolanic reactivity of slag
and fly ash contributes to strength gain at later ages [9,10]. Experimental results in this field visibly
show the crucial effect of the amount of non-clinker constituents in cement on concrete strength
development [12].
However, the examination of early-age volume changes in mass concrete cannot be only limited
to the hydration heat and strength development. As it was emphasized in [4–6] that all material
properties affecting the early-age volume changes and arising stresses in concrete should be considered.
Consequently, a compilation of parameters including rate of hydration heat, thermal conductivity,
heat capacity, thermal dilation, elasticity modulus, tensile strength, and creep with respect to maturity
time were affected strongly by the hardening temperature and should be taken into account. It can
be noticed that many studies addressed the effect of different concrete constituents and describe
properties of early age mass concrete in an independent manner without complex analysis of the
thermo-mechanical behaviour of the real mass structure. Relatively rare works in this area usually
refer to the experimental research of selected properties or validation of experimental tests performed
on laboratory samples [4–6]. If a real structure is analysed, usually part of the input data is based on
theoretical assumptions.
This paper is focused on the overall early age behavior of mass concrete structure made of concrete
with different constituents. In particular, the influence of cement and aggregate type on the formation
of hardening temperature, moisture loss, and stress development in early age concrete has been studied.
The work consists of experimental and numerical study. In the experimental part, six types of cement
with different amount of non-clinker constituents (fly ash and slag) and four types of aggregates (gravel,
basalt, granite, and limestone) have been considered. Special attention was paid to the low clinker
cements with high amount of siliceous fly ash and ground blast furnace slag. Therefore, cements
containing even 68% of non-clinker constituents are used in concrete. The experimental research
covered the determination of thermal (hydration heat, thermal conductivity and capacity, temperature
development), mechanical (elasticity modulus, tensile, and compressive strength), and rheological
(shrinkage and creep) properties of early age concrete with different constituents. First, the obtained
experimental results have been used to validate the applied numerical model. Next, the analysis of
mass foundation slabs has been performed. The main purpose of this analysis was to indicate that
the early-age volume deformation and possible cracking is the result of the concerted action of many
factors relating to thermal and mechanical properties of concrete.

2. Experimental Research

2.1. Materials
Two compositions of concrete mix with different cement and aggregate type were used in the
experiments, as shown in Table 1. Type A concrete was applied for gravel and limestone aggregates
while type B consisted of basalt and granite aggregate.
Six commercial cements with different content of Portland clinker, siliceous fly ash, and/or
granulated blast-furnace slag were used for the experimental tests: Portland slag cement CEM II/B-S
32.5R, Portland-fly ash cement CEM II/B-V 32.5R, slag cement CEM III/A 32.5N-LH/HSR/NA,
composite cement CEM V/A (S-V) 32.5R-LH/HSR/NA, special composite cement with high amount
of non-clinker constituents VLH V/B (S-V), and CEM I 42.5R as a reference cement. The content of
non-clinker constituents in the cement compositions varied from 27.1% to 67.7%, as listed in Table 2.
Materials 2018, 11, 2207 3 of 18

Table 1. Concrete mix composition.

Content, kg/m3
Component
Gravel1 and Limestone Basalt and Granite
Cement (6 types) 300
Water 150
Sand 0/2 mm 583
Aggregate 2/8 mm 427 -
Aggregate 8/16 mm 389 -
Aggregate 16/31.5 mm 544 -
Aggregate 2/8 mm - 427
Aggregate 8/16 mm - 389
Aggregate 16/22.5 mm - 544
1 the aggregate described as a gravel consists mainly of quartz.

Table 2. Composition of cements.

Non-Clinker Constituents, %
Cement Type
Ground blast
Siliceous Fly Ash Minor
Portland Clinker Furnace Slag
(FA) Constituents 1
(GBFS)
CEM I 42.5R 95.7 - - 4.3
CEM II/B-S 32.5R 68.3 27.1 - 4.6
CEM II/B-V 32.5R 65.7 - 29.9 4.4
CEM III/A
41.1 58.9 - -
32,5N-LH/HSR/NA2,3
CEM V/A (S-V)
62.2 18.2 19.6 -
32.5R-LH/HSR/NA2,3
VLH V/B (S-V) 22.5 32.3 34.4 33.3 -
1Limestone with 98.2% of CaCO3 and TOC = 0.07%; 2 HSR—sulphate resistant cement acc. to Polish Standard
PN-B-19707:2013 [13]; 3 NA—low alkaline cement acc. to Polish Standard PN-B-19707:2013 [13].

2.2. Scope and Testing Methods

The research program covered the following tests: physical and mechanical properties of cements,
hydration heat of cements, compressive and tensile strength of concretes, elasticity modulus of
concrete, concrete hardening temperature, thermal conductivity and heat capacity of concretes, concrete
shrinkage, and creep. Some results of this extensive research have been already presented, including
physical and mechanical properties of cements [3,7], hydration heat evolution under adiabatic
conditions and isothermal conditions at 20 ◦ C and 50 ◦ C [3,7], compressive strength development of
concrete with different cements [3,9], hardening temperature of concrete [3], thermal conductivity,
and heat capacity [3]. Details of these tests and their results have been extensively described in previous
works [3,7,9]. Not reported results are: development of compressive strength of concrete with different
aggregate types, tensile strength development, Young’s modulus development, shrinkage and creep
development of concrete with different constituents.
Concrete compressive strength was tested after 1, 2, 7, 28, 56, and 90 days of concrete curing.
Concrete cubes 150 mm × 150 mm × 150 mm were used according to the procedure enclosed in EN
12390-3 [14]. Tensile strength was determined in bending tests after 2 and 28 days. Samples with the
dimensions 150 mm × 150 mm × 700 mm were used, and the testing procedure included in PN-EN
12390-5 [15] has been followed. The modulus of elasticity was determined after 2 and 28 days on
cylindrical samples 150/300 mm in accordance with a test methodology described in ISO 1920-10 [16].
All specimens were stored in the mold for 24 h. Next, the demoulded specimens continued curing at
a temperature of 18 ◦ C (±2 ◦ C) and 90% relative humidity.
Shrinkage strains were measured on cuboidal samples with the dimensions 100 mm × 100 mm ×
500 mm according to the methodology of PN-B-06714-23 [17], after 2, 7, 14, 28, 56, 90, 120, 150, 210,
270, and 360 days of concrete curing. Samples were stored for 24 h in a climate chamber at 18 ◦ C
Materials 11, 2207
2018, 2018,
Materials 11, x FOR PEER REVIEW 4 of 184 of 18

Shrinkage strains were measured on cuboidal samples with the dimensions 100 mm × 100 mm ×
(±2 ◦500
C) and
mm relative
accordinghumidity above 90%.ofNext,
to the methodology the demoulded
PN-B-06714-23 samples
[17], after 2, 7, 14,were subjected
28, 56, to drying
90, 120, 150, 210, in
laboratory conditions at 18 ◦ C (±2◦ C) and 65–75% relative humidity. Similarly, concrete creep strains
270, and 360 days of concrete curing. Samples were stored for 24 h in a climate chamber at 18 °C (±2
were °C)
measured on samples
and relative humidity mm ×
100above 90%. mm ×
100Next, the500 mm. The samples
demoulded same curing
were conditions
subjected towere applied
drying in as
laboratorytests.
for shrinkage conditions
Next, at 18samples
the °C (±2°C)were
and 65–75%
subjected relative humidity. Similarly,
to a compressive concrete
force equal creepofstrains
to 33% the failure
forcewere measured
obtained fromon samples
2-day 100 mm
concrete × 100 mm × strength
compressive 500 mm. The same
tests. Thecuring
creepconditions
strains werewereregistered
applied for
as
90 days. for shrinkage tests. Next, the samples were subjected to a compressive force equal to 33% of the
failure force obtained from 2-day concrete compressive strength tests. The creep strains were
The tests were performed on concretes with the constituents given in Table 1, considering the
registered for 90 days.
different type of aggregate and cement. Tests of concrete with gravel aggregate were made with the
The tests were performed on concretes with the constituents given in Table 1, considering the
use of 6 typestype
different of of
cement.
aggregateCompressive
and cement. andTeststensile strength,
of concrete modulus
with gravel of elasticity,
aggregate were made and
withshrinkage
the
deformations of concretes with basalt, granite, and limestone aggregates were
use of 6 types of cement. Compressive and tensile strength, modulus of elasticity, and shrinkage tested with the
use of 3 types ofofcement:
deformations concretes CEM I 42.5R,
with basalt, CEM
granite, andIII/A 32.5N-LH/HSR/NA,
limestone aggregates were tested and CEM
with V/A
the use of (SV)
32.5R-LH/HSR/NA.
3 types of cement:CreepCEM tests wereCEM
I 42.5R, performed for concrete withand
III/A 32.5N-LH/HSR/NA, gravel
CEMaggregate
V/A (SV) and the six
32.5R-
typesLH/HSR/NA. Creep testsinwere
of cements described Tableperformed
2. for concrete with gravel aggregate and the six types of
cements described in Table 2.
2.3. Test Results
2.3. Test Results
As it has been mentioned in the previous section, some results of this extensive research have
As it has been mentioned in the previous section, some results of this extensive research have
been already presented [3,7,9]. Nevertheless, for completeness of the study and later results of the
been already presented [3,7,9]. Nevertheless, for completeness of the study and later results of the
numerical analysis, some of them are concisely quoted in Figure 1 and Tables 3 and 4.
numerical analysis, some of them are concisely quoted in Figure 1 and Tables 3 and 4.

Figure
Figure 1. Rate
1. Rate of heat
of heat evolution
evolution (a)(a)and
andheat
heat of
of hydration
hydration (b)
(b)under
underisothermal conditions
isothermal at 20at
conditions 20 ◦ C.
°C.

Table 3. Influence
Table of of
3. Influence cement
cementand
andaggregate typeon
aggregate type onconcrete
concretethermal
thermal conductivity
conductivity [3]. [3].

Thermal
ThermalConductivity [W/(m·
Conductivity [W/(m K)]
·K)]
Aggregate
Aggregate Type Type
CementType
Cement Type
Gravel Gravel
Basalt Basalt
Granite Granite
LimestoneLimestone
CEM I 42.5R
CEM I 42.5R 2.43 2.43
1.84 1.84
2.01 2.012.03 2.03
CEM III/A 32.5N-LH/HSR/NA
CEM III/A 2.35 1.70 1.75 1.99
2.35 1.70 1.75 1.99
32.5N-LH/HSR/NA
CEM V/A (S-V) 32.5R-LH/HSR/NA 2.28 1.71 1.98 1.96
CEM V/A (S-V)
2.28 1.71 1.98 1.96
32.5R-LH/HSR/NA
Table 4. Influence of cement and aggregate type on concrete heat capacity [3].

Heatand
Table 4. Influence of cement Capacity [106·
aggregate J/(m3on
type ·K)]concrete heat capacity [3].
Aggregate Type
Cement Type Heat Capacity [106 ·J/(m3 ·K)]
Gravel Basalt Granite Limestone
CEM I 42.5R 1.72
Aggregate1.83Type 1.81 1.71
Cement Type
CEM III/A 32.5N-LH/HSR/NA Gravel Basalt Granite
1.68 1.77 Limestone 1.76
1.74
CEM V/A (S-V)
CEM I32.5R-LH/HSR/NA
42.5R 1.72 1.61
1.83 1.76
1.81 1.771.71 1.67
CEM III/A
1.68 1.77 1.74 1.76
32.5N-LH/HSR/NA
CEM V/A (S-V)
1.61 1.76 1.77 1.67
32.5R-LH/HSR/NA
Materials 2018, 11, 2207 5 of 18

The results of compressive and tensile strength determined for concrete samples with different
types of cement and aggregates are shown in Figures 2 and 3, respectively. The following summary of
these results can be written:
Materials 2018, 11, x FOR PEER REVIEW 5 of 18

• Significantly lower early age compressive and tensile strength are observed in concretes made of
The results of compressive and tensile strength determined for concrete samples with different
cements with non-clinker constituents. These values were lower as the addition of non-clinker
types of cement and aggregates are shown in Figures 2 and 3, respectively. The following summary
constituents
of these resultsincreased. However, the non-clinker constituents in cement favorably improved the
can be written:
compressive strength
• Significantly lower early of concrete after 28 days.
age compressive It is also
and tensile worth
strength arenoting
observedthatin the tensile
concretes strength
made of of
concretes
cements with
withcement containing
non-clinker fly ashThese
constituents. (CEMvalues
II/B-V 32.5R)
were lower is greater than that
as the addition of of concrete with
non-clinker
cementconstituents increased.
containing However,
slag (CEM III/Athe32.5N-LH/HSR/NA),
non-clinker constituentswhile in cementbothfavorably
cementsimproved
revealedthe similar
amountcompressive strength of
of heat evolved concrete
during after 28 days.
hydration (FigureIt is1).also worth noting
It confirms somethat the tensile
other resultsstrength of
[18] indicating
thatconcretes
the use ofwith cementwith
cements containing
fly ashflyimproves
ash (CEMthe II/B-V 32.5R)
tensile is greater
strength ofthan that of concrete with
concrete;
cement containing slag (CEM III/A 32.5N-LH/HSR/NA), while both cements revealed similar
• Concrete containing granite, basalt, or limestone aggregates achieved higher average compressive
amount of heat evolved during hydration (Figure 1). It confirms some other results [18] indicating
strength than concrete with gravel aggregate. The highest average compressive strength was
that the use of cements with fly ash improves the tensile strength of concrete;
obtained
• Concrete for concrete
containing with basalt basalt,
granite, aggregate or and in the early
limestone age period
aggregates it was
achieved even 33.7%
higher average higher
thancompressive
for concrete strength than concrete with gravel aggregate. The highest average compressive was
with gravel aggregate. Concrete with granite or limestone aggregate
characterized
strength was byobtained
early ageforstrength comparable
concrete with to concrete
basalt aggregate with
and in the basalt aggregate;
early age period it was even
• 33.7% higher
Similarly, than fortensile
the average concrete with gravel
strength aggregate.
of concrete Concrete granite,
containing with granite or limestone
basalt, or limestone
aggregate was characterized by early age strength comparable to concrete
aggregates was higher than those with gravel aggregate. In the case of concrete with cement CEM with basalt aggregate;
I• 42.5
Similarly,
R after the average
2 days, theytensile strengthbyof14.7%
were higher concrete containing
to 17.9%. granite,inbasalt,
However, or limestone
the case of CEM III/A
aggregates was higher than those with gravel aggregate. In the case of concrete with cement CEM
32.5N-LH/HSR/NA, the average tensile strength was higher by 45.0 to 145.0% in comparison
I 42.5 R after 2 days, they were higher by 14.7% to 17.9%. However, in the case of CEM III/A
to the gravel aggregate, and in the case of cement CEM V/A (SV) 32.5R -LH/HSR/NA, tensile
32.5N-LH/HSR/NA, the average tensile strength was higher by 45.0 to 145.0% in comparison to
strength was higher
the gravel by 56.4
aggregate, andto in97.3%
the casecompared
of cement to concretes
CEM V/A with (SV) gravel aggregate. After
32.5R -LH/HSR/NA, 28 days,
tensile
the strength
tensile strength
was higher ofbyconcretes with
56.4 to 97.3% basalt, granite,
compared to concretes or with
limestone aggregateAfter
gravel aggregate. was28 higher
days, only
by 23.7%.
the tensile strength of concretes with basalt, granite, or limestone aggregate was higher only by
23.7%.

Figure
Figure 2. 2. Development
Development of concrete
of concrete compressive
compressive strength:
strength: a) gravel,
(a) gravel, b) granite,
(b) granite, c) basalt,
(c) basalt, andand
(d)d)
limestone.
limestone.
Materials 2018, 11, 2207 6 of 18

Materials 2018, 11, x FOR PEER REVIEW 6 of 18

Figure
Figure 3. 3. Development
Development ofof concrete
concrete tensile
tensile strength:
strength: (a)a)gravel,
gravel,(b)
b) granite, (c)
c) basalt,
basalt,and
andd)
(d)limestone
limestone

The
Theobtained
obtainedvalues
values ofofmodulus
modulus of of elasticity
elasticity are
are plotted
plotted in in Figure
Figure 4. The highest
4. The highest early
early age
age
modulusofofelasticity
modulus elasticitywaswasobtained
obtained for
for concrete
concrete withwith Portland
Portland cement
cement CEM CEM I 42.5R.
42.5R. Concretes
Concretes with
with
cementscontaining
cements containingnon-clinker
non-clinkercomponents
componentswere were characterized
characterized by much lower lower modulus
modulusof ofelasticity
elasticity
atatearly
earlyage.
age. Reduction
Reduction of ofthe
themodulus
modulusofofCEM CEM I 42.5R concrete
I 42.5R concrete ranged
ranged from 22.3%
from (CEM
22.3% II/B-S
(CEM 32.5
II/B-S
R) to
32.5 R) 75.4%
to 75.4% (VLH(VLH V/BV/B(S-V) 22.5).
(S-V) Thus,
22.5). the following
Thus, relationship
the following relationshipcan be canstated: the greater
be stated: the
the greater
amount
the amount of ofnon-clinker
non-clinker components
components in inthe
thecement,
cement,the thelower
lowerthe theearly
earlyageage modulus.
modulus. The The elastic
elastic
modulusincreased
modulus increasedsignificantly
significantlyafter
after2828 days
days ofof hardening
hardening for these cements,
cements, however,
however, the theobtained
obtained
valuesare
values arelower
lowerthan
thanininthe
thereference
referenceconcrete
concretewith withPortland
Portlandcement
cementCEM CEMII42.5 42.5R,R,respectively,
respectively,fromfrom7
to712%
to 12% (in the
(in the casecase of concrete
of concrete withwith
CEM CEMII/BS II/BS
32.5R32.5R or CEM
or CEM III/AIII/A 32.5N-LH/HSR/NA)
32.5N-LH/HSR/NA) and
and from
from 19
to1929%
to 29%
(for (for concrete
concrete withwith
CEM CEMV/A V/A(SV)(SV) 32.5R-LH/HSR/NA,CEM
32.5R-LH/HSR/NA, CEMII/BV II/BV 32.5R
32.5R and
and VLH
VLH V/BV/B(SV)(SV)
22.5).Concrete
22.5). Concrete containing
containing granite,
granite, basalt,
basalt, or limestone
or limestone aggregates
aggregates reached reached higher modulus
higher modulus of
of elasticity
elasticity than concretes with gravel aggregate. The highest values are
than concretes with gravel aggregate. The highest values are registered both after 2 days and 28 daysregistered both after 2 days
ofand 28 days
concrete of concrete
curing curingwith
for concrete for concrete with basalt
basalt aggregate aggregate concrete.
concrete.
The results of shrinkage tests are plotted
The results of shrinkage tests are plotted in Figure 5. in Figure 5. It
It has
has been
been found
found that
that the
the use
use of
of cements
cements
with non-clinker main components in the concrete composition reduces the shrinkage ofconcrete.
with non-clinker main components in the concrete composition reduces the shrinkage of concrete.
The following relationship can be also distinguished—the higher the amount of non-clinker main
The following relationship can be also distinguished—the higher the amount of non-clinker main
components is applied; the smaller the concrete shrinkage is. However, this reduction in shrinkage
components is applied; the smaller the concrete shrinkage is. However, this reduction in shrinkage
strains is not in direct proportion to the increase of non-clinker components used in the cement
strains is not in direct proportion to the increase of non-clinker components used in the cement
composition. It is worth adding that the samples had the same maturing conditions, so the differences
composition. It is worth adding that the samples had the same maturing conditions, so the differences
in the shrinkage strains result mainly from different autogenous shrinkage. This is confirmed by
insimilar
the shrinkage
shrinkagestrains
valuesresult mainly from
for concretes with different
CEM II/B-V autogenous
32.5R, CEM shrinkage. This is confirmed
III/A 32.5N-LH/HSR/NA, andby
similar
CEM V/A shrinkage values for concretes which
(SV) 32.5R-LH/HSR/NA, with CEM II/B-V
showed 32.5R,
very CEMamount
similar III/A 32.5N-LH/HSR/NA,
of heat evolved during and
CEM V/A (SV) 32.5R-LH/HSR/NA, which showed very similar amount
hydration (Figure 1). The use of granite, basalt, or limestone as aggregate reduces the shrinkage of heat evolved during
hydration (Figure 1).
strains compared toThe useaggregate.
gravel of granite, The
basalt, or limestone
lowest shrinkage asstrains,
aggregate both reduces the shrinkage
early and long-term,strains
were
compared
registeredtoingravel aggregate.
concrete The aggregate.
with basalt lowest shrinkage strains, both early and long-term, were registered
in concrete with basalt aggregate.
Materials 2018, 11, 2207 7 of 18
Materials 2018,
Materials 2018, 11,
11, xx FOR
FOR PEER
PEER REVIEW
REVIEW 77 of
of 18
18

Figure
Figure 4. Modulus
Modulus
4. Modulus
Figure 4. of elasticity
of of elasticityofof
elasticity ofconcrete:
concrete: a)
concrete: a) gravel,
gravel, b)
gravel, b) granite,
granite,
b)granite, c) c)
c) basalt, andand
basalt,
basalt, and d) limestone.
d) limestone.
d) limestone.

Figure 5. Concrete shrinkage strains: a) gravel, b) granite, c) basalt, and d) limestone.

FigureFigure 5. Concrete
5. Concrete shrinkage
shrinkage strains:
strains: (a)a)gravel,
gravel, (b)
b) granite,
granite,c) (c)
basalt, andand
basalt, d) limestone.
(d) limestone.

The results of the measured creep strains are depicted in Figure 6. The shrinkage has been
subtracted from the creep measurement. Concrete with fly ash cement (CEM II/B-V 32.5 R) is
characterized by the greatest creep strains in the early age period (up to 7 days), as well as in later
Materials 2018, 11, x FOR PEER REVIEW 8 of 18

The results of the measured creep strains are depicted in Figure 6. The shrinkage has 8been
Materials 2018, 11, 2207 of 18
subtracted from the creep measurement. Concrete with fly ash cement (CEM II/B-V 32.5 R) is
characterized by the greatest creep strains in the early age period (up to 7 days), as well as in later
period
period (7–28
(7–28days).
days).TheTheconcrete
concrete creep
creepof of
concrete with
concrete thisthis
with cement
cement is also higher
is also than for
higher than concrete made
for concrete
of Portland cement CEM I 42.5 R (reference concrete). Similar results for concrete
made of Portland cement CEM I 42.5 R (reference concrete). Similar results for concrete with fly ash with fly ash cement
are also mentioned
cement in [10,18]. in
are also mentioned Concrete
[10,18].containing
Concrete slag cement (CEM
containing III/A 32.5N-LH/HSR/NA)
slag cement (CEM III/A 32.5N-
as well as bothasslag
LH/HSR/NA) welland fly ash
as both cement
slag and fly(CEM V/A (SV)
ash cement (CEM 32.5R-LH/HSR/NA)
V/A (SV) 32.5R-LH/HSR/NA) showed theshowedlowest
creep strains among all tested concrete in the early period (up to 7 days). All
the lowest creep strains among all tested concrete in the early period (up to 7 days). All concretes concretes with cements
containing
with cements only granulated
containing blast
only furnace slag
granulated as afurnace
blast non-clinker component
slag as a non-clinker (CEM II/B-S 32.5(CEM
component R andII/B-S
CEM
III/A 32.5 N-LH/HSR/NA) were characterized by lower creep than
32.5 R and CEM III/A 32.5 N-LH/HSR/NA) were characterized by lower creep than the reference the reference concrete, and the
creep strains
concrete, anddecreased
the creep with the increase
strains of slag
decreased in cement
with content.ofSimilar
the increase slag in relationships are mentioned
cement content. Similar
in [19]. In the are
relationships casementioned
of concretes inwith
[19].CEM
In theV/A
case(S-V) 32.5R-LH/HSR/NA
of concretes with CEM V/A and(S-V)
VLH32.5R-LH/HSR/NA
V/B (S-V) 22.5, no
relationship
and VLH V/B was found
(S-V) 22.5,between the amount
no relationship wasof non-clinker
found between component
the amount andof creep strainscomponent
non-clinker level—the
obtained
and creepcreep strains
strains are at aobtained
level—the similar level.
creep Probably,
strains areit at
is the result level.
a similar of theProbably,
opposite effect of result
it is the silica fly
of
ash
the and granulated
opposite blast
effect of furnace
silica fly ashslag ongranulated
and the development of creepslag
blast furnace strains.
on the development of creep
strains.

Figure 6.
Figure 6. Concrete
Concrete creep
creep strains.
strains.

3. Comments on
3. Comments on the
the Experimental
Experimental ResultsResults Regarding
Regarding the the Most
Most Favourable
Favourable Early Early Age
Age Behaviour
Behaviour
of Mass Concrete
of Mass Concrete
The best concrete for a massive structure would be the concrete with:
The best concrete for a massive structure would be the concrete with:
• smallsmallheat
heatevolved
evolvedduringduringcement
cement hydration
hydration (resulting
(resulting in in low
low hardening
hardening temperature),
temperature),
•• highhighthermal
thermalconductivity
conductivity(resulting
(resulting inin fast
fast heat
heat dissipation
dissipation from from the
the structure),
structure),
•• highhighheat
heatcapacity
capacity(resulting
(resulting in in lower
lower hardening
hardening temperature),
temperature),
•• small
smallcoefficient
coefficient of of thermal
thermal expansion
expansion (resulting
(resulting inin lower
lower deformation),
deformation),
•• small shrinkage (lowering the total early age
small shrinkage (lowering the total early age deformation), deformation),
•• small
smallmodulus
modulusof ofelasticity
elasticityand andhighhighcreep
creep (decreasing
(decreasing of of the
the arising
arising stresses),
stresses),
• high tensile strength (increasing the early age cracking resistance).
• high tensile strength (increasing the early age cracking resistance).
Referring the obtained experimental results to the above-mentioned beneficial properties, the
Referring
following the obtained
comments experimental results to the above-mentioned beneficial properties,
can be written:
the
• following
the use of comments
cements with canabe written:
large amount of non-clinker components in mass concrete is beneficial
• due to reduced
the use of cements hydration
with a largeheatamount
which decreases the hardening
of non-clinker components temperature and to
in mass concrete a smaller
is beneficial
elastic
due to reduced hydration heat which decreases the hardening temperature and to a smallerofelastic
modulus which decreases the arising stresses; nevertheless, the tensile strength such
concrete
modulusiswhichmuchdecreases
lower and theconsequently the early
arising stresses; age cracking
nevertheless, resistance
the tensile decreases,
strength of such concrete is
• the muchuselower
of cements with a large number
and consequently the earlyofagenon-clinker
cracking components results also in lower shrinkage
resistance decreases,
• whichthe useisofpreferred
cements in mass
with concrete,
a large number of non-clinker components results also in lower shrinkage
• greater
which is early age creep
preferred is beneficial
in mass concrete,in mass concrete as it reduces early age stresses, therefore, in
• this regard,
greater earlyit age
is better
creepto is use cements
beneficial in containing
mass concrete silica
as fly ash rather
it reduces than
early agecements
stresses,containing
therefore,
ground blast furnace
in this regard, slag to
it is better which showed containing
use cements lower creep, silica fly ash rather than cements containing
• the effectblast
ground of aggregate
furnace slag is particularly
which showed complex.
lowerConsidering
creep, thermal conductivity, heat capacity
• and modulus of elasticity concrete with gravel
the effect of aggregate is particularly complex. Considering thermalaggregate has theconductivity,
best propertiesheatfor mass
capacity
and modulus of elasticity concrete with gravel aggregate has the best properties for mass concrete,
Materials 2018, 11, 2207 9 of 18

but simultaneously, the thermal dilation of concrete with this aggregate is the largest [20,21].
In contrast, concrete containing granite, basalt, or limestone aggregates reached higher modulus of
elasticity and lower thermal conductivity and heat capacity than concretes with gravel aggregate,
but the thermal dilation of concrete, tensile strength, and shrinkage are more beneficial.

4. Thermo-Mechanical Analysis

4.1. Numerical Model

The original numerical model has been developed for the 3D analysis of early age concrete
and it can be applied for reinforced concrete structures as foundation slabs and blocks, tank walls,
and bridge abutments. The complex analysis of the mentioned structures is run in two stages. In the
first stage, temperature and moisture development in time are determined. The original feature of
this stage is the application of coupled equations of thermodiffusion. Next, thermal-shrinkage strains
are calculated, and these results are used as an input for computation of stresses in the second stage
of analysis. In analysis of early age stresses, the viscoelastic material model of ageing concrete is
applied. The influence of elevated curing temperatures on concrete mechanical properties as well on
the development of hydration heat is considered by introducing equivalent concrete age instead of the
real age of concrete [7,22,23].
The model considers soil–structure interaction and appropriate technological conditions like
layered casting, environmental conditions like ambient temperature and humidity, different initial
temperature of the concrete layers, and the time of formwork removal. The full thermo-mechanical
concrete properties are also considered. For practical application, original computers programs have
been developed: TEMWIL for thermal–moisture analysis and MAFEM for stress analysis. Detailed
description of the model and computer programs TEMWIL and MAFEM can be found in [22–24].

4.2. Validation of the Model

First, experimental results have been computer program TEMWIL used to validate the numerical
model. Validation of the model covered the determination of hardening temperature of concrete and
shrinkage and creep strains. Some considerations on the correct description of the development of
mechanical properties and proper value of the activation energy of concretes with different constituents
have been enclosed in [9,25] and were used in the analysis.
The results of numerical calculations and experimentally registered temperature of cubic
specimens 40 cm × 40 cm × 40 cm [3] are depicted in Figure 7. In all cases, an acceptable agreement
was achieved for temperature development in time. That proves the proper parameters of the model
assumed in thermal analysis. The detailed values of thermal parameters are summarized in Table 5.
In the validation of shrinkage tests, the best agreement between the results of calculations and the
experiment was obtained for the following values assumed in the numerical model: moisture transfer
coefficient equal to 2.78 × 10−8 m/s, coefficient of moisture dilatation αw = 0.001/0.0009/0.0006/0.0008
for gravel/granite/basalt/limestone. The other parameters were adopted in accordance with Table 5.
Figure 8 compares experimentally-measured shrinkage with numerical results. It is worth noting that
the differences in the shrinkage deformations for concretes with different cements result from the
different autogenous shrinkage of these concretes, because the external conditions were the same for
all samples. This effect is possible to simulate in the computer programme TEMWIL as the coupled
equations of heat and mass transfer are used.
A concrete creep test was also simulated. The calculations considered the development of
the mechanical properties of concretes determined during experimental investigations and other
experimental data. Good agreement between the results of numerical calculations and the results of
experimental tests can be noticed in Figure 9.
Materials 11, 2207
2018, 2018,
Materials 11, x FOR PEER REVIEW 10 of 18
10 of 18

Table 5. Basic data for numerical validation.

Table 5. Basic data for numerical validation.
Properties/Coefficient Value
Properties/Coefficient
Concrete composition Value 1
acc. to Table
Concrete composition 0 acc. to Table 3, with consideration
acc. to Table 1 of hydration rate α:
Thermal conductivity
0 (2 −  ) [26]
λ0 acc. to Table =3,with consideration of hydration
Thermal conductivity
rate α: λ = λ (2 − α) [26]
0
cb 0 acc. to Table 4, with consideration of hydration rate α:
Specific heat cb0 acc. to Table 4, with consideration of hydration
Specific heat c = c (1.25 − 0.25) [26]
rateb α: bc0b = cb0 (1.25 − 0.25α) [26]
Coefficient representing the
Coefficient representing the influence of the −5 2 K/s
influence of the moisture
moisture concentration on heat transfer
9.375
9.375 ×× 1010
−5 m2mK/s
concentration on heat transfer
Thermal transfer coefficient 2
0.248 W/(m K)–all surfaces (150 mm insulation [3])
Thermal transfer coefficient 0.248 W/(m2K)–all surfaces (150 mm insulation [3])
on the basis of Figure 1 and results reported in [7,25],
Heat of hydration on the basiswithof Figure 1 and results
the consideration reported
of concrete in [7,25], with
hardening
Heat of hydration the consideration
temperatureof byconcrete
means ofhardening
an equivalent temperature
concrete ageby
Water-cement proportionality means of an equivalent
0.3 × 10−9 m concrete
3 /J age
Water-cement proportionality 0.3 × 10−9 m3/J
Moisture diffusion on the basis of Hancox’s equation [22]
Moisture diffusion on the basis of Hancox’s equation [22]
Thermal moisture diffusion 10−112 m2 /(sK)
2 × −11
Thermal moisture diffusion 2 × 10 m /(sK)
−8 m/s–all surfaces (150 mm insulation [3])
MoistureMoisture
transfertransfer coefficient
coefficient 0.01·× 10·×
0.01 −8 10
m/s–all surfaces (150 mm insulation [3])
on theonbasis
the basis of experimental
of experimental tests
tests and andconsiderations
considerations
reported in [9], with the consideration of concrete
Mechanical properties reported in [9], with the consideration of concrete
Mechanical properties hardening temperature by means of an equivalent
hardening temperature by concrete means of agean equivalent concrete
age
on the basis of experimental tests reported in [25]:
on the44,178
basis J/mol
of experimental
- CEM I 42.5R, tests reported
40,576 J/mol in [25]:
- CEM
44,178II/B-S
J/mol32.5R,
- CEM I 42.5R,
43,935 J/mol40,576
- CEMJ/molII/B-V- 32.5R,
CEM 35,154
II/B-S
Activation energy
Activation energy 32.5R, 43,935J/mol - CEM
J/mol - CEMIII/A 32.5N-LH/HSR/NA,
II/B-V 32.5R, 35,154 J/mol 34,459 - CEM
J/mol - CEM V/A (S-V) 32.5R-LH/HSR/NA, 31,117
III/A 32.5N-LH/HSR/NA, 34,459 J/mol - CEM V/A (S-V)
J/mol - VLH V/B (S-V) 22.5
32.5R-LH/HSR/NA, 31,117 J/mol - VLH V/B (S-V) 22.5

Figure
Figure 7. 7. Experimental
Experimental and and numerical
numerical resultsofofhardening
results hardening temperature:
temperature: a) (a)
gravel, b) granite,
gravel, c)
(b) granite, (c)
basalt, and d) limestone.
basalt, and (d) limestone.
Materials 2018, 11, 2207 11 of 18

Materials 2018, 11, x FOR PEER REVIEW 11 of 18

Materials 2018, 11, x FOR PEER REVIEW 11 of 18

Figure8.8.Experimental
Figure Experimentaland
andnumerical
numericalresults
resultsofofshrinkage
shrinkagestrains.
strains.
Figure 8. Experimental and numerical results of shrinkage strains.

Figure 9. Experimental and numerical results of creep strain.

Figure 9.
Figure 9. Experimental
Experimental and
and numerical
numerical results
results of
of creep
creep strain.
strain.
4.3. Analysis of Massive Foundation Slabs
4.3.
4.3. Analysis
Analysis of of Massive
Massive Foundation
Foundation Slabs
Slabs
4.3.1. Object and Scope of Analysis
4.3.1.
4.3.1. Object
Object and
and Scope
Scope of
of Analysis
Analysis
Finally, the complex influence of concrete constituents on the thermo-mechanical behavior has
beenFinally, the
the complex
investigated
Finally, in massinfluence
complex of
of concrete
foundation
influence constituents
slabs with
concrete on
on the
the thermo-mechanical
the base dimensions
constituents 20 m × 20 m and behavior
thermo-mechanical 4behavior has
m thickness.
has
been investigated
The investigated
been finite element in mass
in mesh foundation slabs
of the slab slabs
mass foundation with
is shown
withinthe base
theFiguredimensions
10. Only a 20
base dimensions 20 m ×
quarter20 m and
ofmthe
m × 20 and 4
slab m thickness.
4 misthickness.
modelled
because
The finiteofelement
its symmetry.
mesh ofThethedata
slab for the numerical
is shown in Figure study was taken
10. Only on the
a quarter baseslab
of the of experimental
is modelled
tests described
because in former The
of its symmetry. sections.
data The same
for the data given
numerical in Table
study 5 was applied
was taken withofthe
on the base exception of
experimental
tests described in former sections. The same data given in Table 5 was applied with the exception of
Materials 2018, 11, 2207 12 of 18

The finite element mesh of the slab is shown in Figure 10. Only a quarter of the slab is modelled
because of its symmetry. The data for the numerical study was taken on the base of experimental
Materials 2018, 11, x FOR PEER REVIEW 12 of 18
tests described in former sections. The same data given in Table 5 was applied with the exception
of 2 K) (top surface without any protection) and
thethe thermal
thermal transfer
transfer coefficient
coefficient equal
equal to W/(m
to 6.0 6.0 W/(m
2K) (top surface without any protection) and equal
equal to 3.52K)W/(m 2
to 3.5 W/(m (sideK) (side surfaces
surfaces with plywood).
with plywood). Similarly,transfer
Similarly, moisture moisture transfer was
coefficient coefficient
assumedwasto
− 8 − 8
assumed
2.78 × 10−8tom/s2.78at× 10surface
top m/s and
at top
to surface
0.18 × 10and to at
−8 m/s 0.18
side× surfaces.
10 m/sAmbient
at side surfaces.
and initialAmbient and
temperature
initial
of the temperature
slab were taken of the slab
as 20 °C.were
All taken as 20 ◦ C. cases
computational All computational cases6.
are listed in Table are listed in Table 6.

Figure 10. Assumed

Figure 10. Assumed finite
finite element
element mesh
mesh in
in the
the slab.
slab.

Table 6. Analyzed cases.

Table 6. Analyzed cases.
Slab Thickness Analyzed Case Analyzed Results for All Cases
Slab Analyzed Results for All
Analyzed Case
gravel aggregate and 6 types of cement
Thickness 4m Cases
Temperature development
used in concrete mix for each slab
4m gravel aggregate and 6 types of cement used in Moisture loss development
Temperature development
basalt/granite/limestone aggregate and Development ofdevelopment
tensile
concrete CEM
mix for each slab
III used in concrete mix for each slab Moisture loss
stresses at top surface of slabs
basalt/granite/limestone aggregate and CEM III Development of tensile
used in concrete
4.3.2. Results and Discussion mix for each slab stresses at top surface of slabs

4.3.2.InResults
thick foundation slabs, the self-induced stresses resulting from non-uniform volume changes
and Discussion
in a cross section are predominant. These stresses have a distinctive distribution during concrete
In thick foundation slabs, the self-induced stresses resulting from non-uniform volume changes
hardening. In the heating phase, the tensile stresses exist in surface layers, while compressive stresses
in a cross section are predominant. These stresses have a distinctive distribution during concrete
are observed inside the slab. Next, an inversion of the stresses occurs in the cooling phase. In this
hardening. In the heating phase, the tensile stresses exist in surface layers, while compressive stresses
phase, the tensile stresses are observed in the core of the slab and compressive stresses exist in the
are observed inside the slab. Next, an inversion of the stresses occurs in the cooling phase. In this
surface layers [23]. Thus, in foundation slabs, the early age cracks are usually observed on the top
phase, the tensile stresses are observed in the core of the slab and compressive stresses exist in the
surface in the heating phase, because of the tensile stresses existing in surface zones of the slab. That is
surface layers [23]. Thus, in foundation slabs, the early age cracks are usually observed on the top
why the results regarding the development of stresses has been limited to the presentation of tensile
surface in the heating phase, because of the tensile stresses existing in surface zones of the slab. That
stresses on the surface of the slabs. Similarly, in analysis of hardening temperature, the results have
is why the results regarding the development of stresses has been limited to the presentation of tensile
been discussed in relation to the recommended maximum temperature difference between the core
stresses on the surface of the slabs. Similarly, in analysis of hardening temperature, the results have
and the surface of a concrete structure equal to 15–20 ◦ C [1,2,19].
been discussed in relation to the recommended maximum temperature difference between the core
andInfluence
4.4. the surface of a concrete
of cement type structure equal to 15–20 °C [1,2,19].
Influence of cement type
The results of numerical analysis of massive foundation slabs with different cement types are
The results
presented of numerical
in Figures 11 and 12. analysis of massive
The following foundation
remarks slabs
can be made with
from different
the cement types
above-mentioned are
figures:
presented in Figures 11 and 12. The following remarks can be made from the above-mentioned
-figures:
In case of cements: CEM II/B-V 32.5R and CEM II/B-S 32.5 R, which contain silica fly ash or
- In granulated blast furnace
case of cements: CEM slag of similar
II/B-V amount
32.5R and CEM(about
II/B-S 30%), thewhich
32.5 R, highercontain
performance of fly
silica fly ashash
or
was observed. Slabs with CEM II/BV 32.5R cement were characterized by 14.2 ◦ C lower core
granulated blast furnace slag of similar amount (about 30%), the higher performance of fly ash
temperature
was observed.inSlabs
relation
withtoCEM
CEM II/BV
II/ BS32.5R
32.5Rcement
cement.were
The characterized
temperature difference
by 14.2 °Cbetween the
lower core
core and theinsurface ◦
temperature relationoftothe
CEM slabs made
II/ BS of cement.
32.5R CEM II/BV 32.5R did notdifference
The temperature exceed 20 C despite
between the
the core
considerable thickness of the slab. Also, tensile stresses at the top surface
and the surface of the slabs made of CEM II/BV 32.5R did not exceed 20 °C despite the are much lower than
considerable thickness of the slab. Also, tensile stresses at the top surface are much lower than
the current tensile strength of concrete. It is worth noting that concrete with CEM II/B-V 32.5R
has a low modulus of elasticity and high creep, which is beyond the low hydration heat, also the
reason for its low tensile stresses in comparison to CEM II/BS 32.5R,
Materials 2018, 11, 2207 13 of 18

the current tensile strength of concrete. It is worth noting that concrete with CEM II/B-V 32.5R
has a low modulus of elasticity and high creep, which is beyond the low hydration heat, also the
reason for its low tensile stresses in comparison to CEM II/BS 32.5R,
- The use of CEM III/A 32.5N-LH/HSR/NA cement with a slag content of 58.9% does not
guarantee early age cracking resistance of the slabs. In this case, the tensile stresses are greater
than the
Materials 2018,tensile
11, x FORstrength,
PEER REVIEW 13 of 18

- In the case of using CEM V/A (SV) 32.5R-LH/HSR/NA (the content of non-clinker components
- The use of CEM III/A 32.5N-LH/HSR/NA cement with a slag content of 58.9% does not guarantee
was 37.8%), a decrease in maximum temperature and center-surface temperature difference was
early age cracking resistance of the slabs. In this case, the tensile stresses are greater than the
observed in the foundation slab. The mentioned values were at a level similar to that of CEM
tensile strength,
- III/A
In the32.5N-LH/HSR/NA,
case of using CEM V/A in (SV)
which the addition of the
32.5R-LH/HSR/NA (the non-clinker component
content of non-clinker is much higher
components
(58.9%). Tensile
was 37.8%), stressesinare
a decrease relatively
maximum high in relation
temperature to the tensile
and center-surface strength,
temperature but theywas
difference have not
observed
exceeded it, in the foundation slab. The mentioned values were at a level similar to that of CEM
- TheIII/A 32.5N-LH/HSR/NA,
foundation in which
slabs of concrete the addition
containing VLH ofV/B
the non-clinker
(S-V) 22.5 were component is muchby
characterized higher
the lowest
(58.9%). Tensile stresses are relatively high in relation to the tensile
maximum center temperature as well as the lowest center-surface temperature difference. strength, but they have not This is
exceeded it,
due to the lowest heat of hydration of VLH V/B (S-V) 22.5 of all the analyzed cements. The values
- The foundation slabs of concrete containing VLH V/B (S-V) 22.5 were characterized by the lowest
of the surface stress were halved in comparison to concrete with cement CEM I 42.5R. This was
maximum center temperature as well as the lowest center-surface temperature difference. This is
also
duedue to the
to the lowest muchheatlower modulus
of hydration of elasticity
of VLH V/B (S-V)of concrete
22.5 of all thewith this cement.
analyzed cements. The values
- Theof the surface stress were halved in comparison to concrete with cement CEM affects
increase in the amount of non-clinker main components in the cement I 42.5R.theThisdelay
was of the
maximum
also due to temperature
the much lower andmodulus
maximum tensile strength
of elasticity of concreteoccurrence. The following relationship is
with this cement.
The increase
- visible: the higherin thethe amount
contentof non-clinker main components
of the non-clinker component in theincement affects the delay
the cement, greaterof the
the delay
maximum temperature and maximum tensile strength occurrence.
in comparison to the reference concrete with Portland cement CEM I 42.5R. At the same The following relationship is time,
visible: the higher the content of the non-clinker component in the
it was not noticed that the type of non-clinker component influenced the discussed delay. cement, the greater the delay
in comparison to the reference concrete with Portland cement CEM I 42.5R. At the same time, it
- It was observed that the type of cement affects the moisture content of concrete in the analyzed
was not noticed that the type of non-clinker component influenced the discussed delay.
- period. These differences
It was observed that the type areofthe result
cement of the
affects thevarying
moistureamount
content of ofconcrete
water bound in the cement
in the analyzed
hydration
period. These differences are the result of the varying amount of water bound in the released.
process, which is a direct consequence of the different amount of heat cement It is
worth noting
hydration that significantly
process, which is a direct greater moisture
consequence of loss occurred
the different on theofsurface
amount of theItanalyzed
heat released. is
slabs,
worth which
noting is that
consistent with greater
significantly the work [27], which
moisture states that
loss occurred considerable
on the surface of themoisture
analyzed loss and
slabs, which
shrinkage appear is consistent with the
in the surface work
layer [27], which
of mass concretestates
butthat
do considerable
not significantlymoisture
affectloss
theand
moisture
shrinkage appear in
content inside this structure. the surface layer of mass concrete but do not significantly affect the moisture
content inside this structure.

Figure Figure
11. The11. The results
results of numerical
of numerical studystudy of slabs
of slabs with different
with different cement
cement type:type: a) temperature
(a) temperature development,
development, b) tensile stress and strength development.
(b) tensile stress and strength development.
Materials 2018, 11, 2207 14 of 18
Materials 2018, 11, x FOR PEER REVIEW 14 of 18

Figure 12. The results of numerical study of slabs with different different cement
cement type:
type: (a)
a) maximum
maximum hardening
hardening
temperature
temperature TTmax
max,, b)
(b)maximum
maximumcenter-top
center-top surface
surface temperature difference,
difference,(c)
c) time of occurrence
occurrence TTmax
max,,

d) time of occurrence of maximum tensile stress s , e) moisture content at the center of

(d) time of occurrence of maximum tensile stress smax , (e) moisture content at the center of the slabs,
max the slabs, f)
moisture
(f) content
moisture at the
content toptop
at the of the slabs.
of the slabs.

4.5. Influence
Influence of aggregate
of aggregate type type
The results
The results ofofnumerical
numericalteststestsarearepresented
presented ininFigures
Figures 13 13
andand14.14.
First of all,
First it can
of all, be noticed
it can that
be noticed
the type of aggregate does not significantly affect the temperature distribution
that the type of aggregate does not significantly affect the temperature distribution in the foundation in the foundation slab.
The
slab.effect of theof
The effect aggregate type on
the aggregate theon
type maximum
the maximumtemperature of the core
temperature and
of the coretheand
surface is not greater
the surface is not
than 3 ◦ C. Small differences in the maximum temperature reflect small differences in the heat capacity
greater than 3 °C. Small differences in the maximum temperature reflect small differences in the heat
of concretes
capacity with thesewith
of concretes aggregates. Similar results
these aggregates. were results
Similar obtained during
were experimental
obtained during research.
experimental
The difference in temperature between the core and the surface exceeded 20 ◦ C in all slabs.
research.
Nevertheless, only in
The difference incase of concrete
temperature with gravel
between the coreaggregate
and the were stresses
surface exceeding
exceeded 20 °C theinmaximum
all slabs.
concrete tensile
Nevertheless, strength
only in caseobserved.
of concrete Thewithreason
gravel is the low tensile
aggregate werestrength
stresses and higherthe
exceeding coefficient
maximum of
thermal expansion of concrete (resulting from the presence of quartz in the gravel
concrete tensile strength observed. The reason is the low tensile strength and higher coefficient of aggregate) compared
to concretes
thermal with other
expansion aggregates.
of concrete In this from
(resulting case, the
the lower
presence elastic
of modulus
quartz inofthe concrete
gravel with gravel
aggregate)
aggregate did
compared not balance
to concretes withthe larger
other thermal deformations.
aggregates. In this case, theThus, lowerthe complex
elastic moduluseffectofofconcrete
the concrete
with
gravel aggregate did not balance the larger thermal deformations. Thus, the complex effect ofhere.
mix composition on the thermo-mechanical behavior of massive structures is particularly visible the
The mix
concrete typecomposition
of aggregate onused in the concrete composition
the thermo-mechanical behavior ofaffectsmassivethestructures
time of the occurrence
is particularly
of maximum
visible here. temperatures and maximum tensile stresses. The shortest time was obtained for
concrete
The with
type gravel aggregate.
of aggregate usedIninthe thecase of concretes
concrete with limestone,
composition affects thegranite,
time of or the
basalt aggregates,
occurrence of
the maximum
maximum temperature
temperatures andoccurred
maximum 12,tensile
24, and 36 h later,
stresses. The respectively,
shortest time in was comparison
obtained for to concrete
concrete
with
with gravel
gravel aggregate.
aggregate. This is due
In the casetoofthe differentwith
concretes thermal conductivity
limestone, of these
granite, concretes.
or basalt Similarly,
aggregates, the
the maximum tensile stresses occurred 24, 36, and 48 h later than for concrete
maximum temperature occurred 12, 24, and 36 h later, respectively, in comparison to concrete with with gravel aggregate.
gravelThe use of gravel
aggregate. This aggregate
is due to thein the concrete
different is the conductivity
thermal least favorable ofdue
theseto concretes.
the valuesSimilarly,
of maximum the
stresses in the analyzed foundation slabs, and due to the relatively short
maximum tensile stresses occurred 24, 36, and 48 h later than for concrete with gravel aggregate. time of their occurrence.
The use of gravel aggregate in the concrete is the least favorable due to the values of maximum
stresses in the analyzed foundation slabs, and due to the relatively short time of their occurrence.
Materials2018,
Materials 2018,11,
11,2207
x FOR PEER REVIEW 15 of 18
15 of 18

Materials 2018, 11, x FOR PEER REVIEW 15 of 18

Regarding the mentioned properties, the granite aggregate in the concrete composition seems to be
Regarding
the the mentioned
most optimal.
Regarding the properties,
The analyzed
mentioned the
thegranite
foundation
properties, aggregate
slabs
granite made in
aggregate ofinthe
theconcrete
this concretecomposition
concrete seems
seemsto
were characterized
composition tobe
be the
by the
most
lowestoptimal.
maximum
the most The analyzed foundation
temperature
optimal. The slabs made
and tensileslabs
analyzed foundation of
stresses.this concrete
made Occurrence were characterized
of maximum
of this concrete by the
temperature
were characterized lowest
by theand
maximum
maximum temperature
lowest maximum and tensile
stresses temperature
is also andstresses.
delayed. The use
tensile Occurrence
of basalt
stresses. ofand
maximum
Occurrence temperature
limestone
of maximum aggregatesand
is maximum
alsoand
temperature more
stresses is also
advantageous delayed.
than The
gravel use of
aggregate.basalt and limestone aggregates is also more
maximum stresses is also delayed. The use of basalt and limestone aggregates is also more advantageous than
gravel aggregate.
advantageous than gravel aggregate.

Figure 13. The results of numerical study of slabs with different aggregate type: a) temperature
Figure13.13.The
Figure Theresults
resultsofofnumerical
numerical study
study of slabs with
with different
different aggregate
aggregatetype:
type:a)(a)
temperature
temperature
development, b) tensile stress and strength development
development,(b)
development, b) tensile
tensile stress
stress and
and strength
strengthdevelopment
development.

Figure14.
Figure results of
14. The results of numerical
numerical study
study of
of slabs
slabs with
with different
differentaggregate
aggregatetype:
type:a)(a)
maximum
maximum
Figure14. The results of numerical study of slabs with different aggregate type: a) maximum
hardeningtemperature
hardening temperatureTTmax b) maximum
max, (b) maximumcenter-top
center-topsurface
surfacetemperature
temperaturedifference, c) (c)
difference, time of of
time
hardening temperature Tmax, b) maximum center-top surface temperature difference, c) time of
occurrenceTmax
occurrence Tmax,, and
and (d)
d) time
timeofofoccurrence
occurrenceofof
maximum
maximum tensile stress
tensile smaxs.max .
stress
occurrence Tmax, and d) time of occurrence of maximum tensile stress smax.
5. 5.
Conclusions
Conclusions
5. Conclusions
This
Thisresearch
researchpresents
presentsthe theresults
results of experimentaland
of experimental andnumerical
numericalstudies studies onon thethe overall
overall early-age
early-age
This
properties research
and presents
behavior of the
mass results
concrete of experimental
structures. and
Since numerical
concrete
properties and behavior of mass concrete structures. Since concrete is a heterogeneous material is a studies on
heterogeneous the overall
material early-age
whose
properties
behaviour and behavior
dependsdepends
whose behaviour of mass
on the properties concrete structures.
of its constituents,
on the properties Since
of its constituents, concrete
the cement
the cement is a
and and heterogeneous
aggregate type
aggregate material
hashas
type been
whose
studied. behaviour
Although
been studied. depends
in some
Although on the
works
in some properties
[28,29]
works the of
[28,29] theitsproperties
constituents,
properties of concrete the are
of concrete cement
tested
are and
testedandaggregate
anddescribed
describedtype has
atatthe
been studied. Although
the microstructure
microstructure level,level, in
thesome
the research works
research
results [28,29]
results the properties
published
published in this
in this of concrete
article
article may may are
also
also bebe tested and described at
useful.
useful.
the microstructure
Theuse
The useofofcementslevel,with
cements the research
with non-clinker
non-clinker results
main
main published
components
components in this article
ininmass
mass may also
concrete,
concrete, be useful.
especially
especiallywith
witha large
a large
number
number Theofuse
ofthemof cements
them (>35%),with
(>35%), reduces non-clinker
reduces the maximum
the maximummain components
hardening
hardeningtemperaturein mass concrete,
temperature and
and especially
generated
generated withand
stresses
stressesa large
and
delays
number the
of time
them of their
(>35%), occurrence.
reduces The
the presented
maximum results
hardening show that the
temperature
delays the time of their occurrence. The presented results show that the use of cements with addition use of
and cements with
generated addition
stresses and
of
delays siliceous
the timeflyofash
theirin concrete
occurrence. is more
The advantageous
presented results than
show cements
that
of siliceous fly ash in concrete is more advantageous than cements containing granulated blast furnace containing
the use of granulated
cements with blast
addition
offurnace
slag siliceous
in mass slag in mass
fly ash in
concrete. concrete.
concrete
Experimental Experimental
istests
more and tests and numerical
advantageous
numerical ofanalysis
than cements
analysis foundationof foundation
containing slabs
granulated
slabs made madeblast
of concretes
of concretes
furnace slag containing
in masswith cements
concrete. with silica
Experimental fly ash showed that the maximum temperature and tensile
containing cements silica fly ash showedtests thatand numerical temperature
the maximum analysis of foundation
and tensileslabs
stressmade
are
stress
of concretesare lower, which results from the lower hydration heat of these cements, lower elasticity
lower, whichcontaining
results from cements with silica
the lower fly ash
hydration showed
heat of thesethatcements,
the maximum lowertemperature and tensile
elasticity modulus of
modulus
stress arelessof concrete,
lower, whichand less shrinkage,
results from andlower
the greater creep compared
hydration heatwith to concretes
of these cements, with
lower cements
elasticity
concrete, shrinkage, greater creep compared to concretes cements containing granulated
containing granulated blast furnace slag.
modulus
blast furnace of slag.
concrete, less shrinkage, and greater creep compared to concretes with cements
containing granulated blast furnace slag.
Materials 2018, 11, 2207 16 of 18

It should be noted that the conclusions regarding the benefits of using cement with non-clinker
main components in massive structures are not new. However, they were mainly related to limiting
the hardening temperature of concrete. Furthermore, the composition of the concrete was usually
determined in terms of limiting the maximum temperature and ensuring adequate compressive
strength of the concrete without a comprehensive analysis of all its properties. Similarly, other
beneficial properties of such cements (low shrinkage, high creep, low modulus of elasticity) are not
raised very often.
Based on the analyses and experimental tests, the tested cements may be ordered from the most
to the least favorable based on early thermal behaviour in mass concrete: VLH V/B (SV) 22.5; CEM
II/BV 32.5R; CEM V/A (SV) 32.5R-LH/HSR/NA; CEM III/A 32.5N-LH/HSR/NA; CEM II/BS 32.5R;
CEM I 42.5R. Therefore, considering the type of cement, much better behaviour of mass concrete has
been noted for cements with fly ash and composite cements containing both fly ash and slags than
cements with slag only.
It is also worth mentioning that curing temperature exceeding 65–70 ◦ C can lead to delayed
ettringite formation (DEF) and consequently to possible later damage of the concrete. In this respect,
as it was indicated in [30,31] that the use of silica fly ash or granulated blast furnace slag as a cement
component has the following additional advantages:

• silica fly ash above 15% reduces DEF, and with content above 30%, DEF does not occur,
• granulated blast furnace slag in the amount of 17.5% limits DEF and is preventative above 25–30%.

The significant influence of aggregate type on the stress development in the mass structures has
also been highlighted. The conducted analysis revealed that the gravel aggregate, consisting mainly of
quartz, can be considered as the worst aggregate, while granite aggregate may be treated as the best.
Therefore, the proper selection of aggregate is also important for reducing early-age cracking in mass
concrete. Simultaneously, the effect of the aggregate is very complex and all thermal and mechanical
properties should be carefully considered.
Thus, the results confirm that the early-age volume deformation and possible cracking is the result
of the concerted action of thermal and mechanical properties of concrete whose depend primarily on
the concrete constituents.
Finally, despite the fact that the research program was relatively extensive, it requires continuation
by others in the field of tensile creep testing, as concrete exhibits higher creep under tension [32,33]
and the early-age tensile stresses determined under the assumption of symmetrical creep may be
overestimated. Similarly, the tensile strength of mass concrete should be investigated due to the
published research indicating its reduction in large structures [34].

Author Contributions: Conceptualization, B.K.; Data curation, M.B.; Formal analysis, A.Ż.; Investigation, B.K. and
M.B.; Methodology, Z.G.; Supervision, Z.G.; Validation, B.K.; Writing–original draft, B.K. and A.Ż.; Writing–review
& editing, B.K. and Z.G.
Funding: This research received no external funding.
Acknowledgments: The co-author of the paper, Barbara Klemczak, is a scholar under the Projects:
03/060/RGJ17/0038 funded by Rector of Silesian University of Technology and BK-237/RB-6/2018 funded
by Silesian University of Technology.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. ACI Committee 207. Guide to Mass Concrete; American Concrete Institute: Farmington Hills, MI, USA, 2006.
2. ACI Committee 207. Report on Thermal and Volume Change Effects on Cracking of Mass Concrete; American
Concrete Institute: Farmington Hills, MI, USA, 2007.
3. Batog, M.; Giergiczny, Z. Influence of mass concrete constituents on its properties. Construct. Build. Mater.
2017, 146, 221–230. [CrossRef]
Materials 2018, 11, 2207 17 of 18

4. Pane, I.; Hansen, W. Investigation on key properties controlling early-age stress development of blended
cement concrete. Cem. Concr. Res. 2008, 38, 1325–1335. [CrossRef]
5. Zhou, W.; Feng, C.; Liu, X.; Liu, S.; Zhang, C.; Yuan, W. Contrastive Numerical Investigations on
Thermo-Structural Behaviors in Mass Concrete with Various Cements. Materials 2016, 9, 1–19. [CrossRef]
[PubMed]
6. J˛edrzejewska, A.; Benboudjema, F.; Lacarriere, L.; Azenha, M.; Schlicke, D.; Dal Pont, S.; Delaplace, A.;
Granja, J.; Hajkova, K.; Heinrich, P.J.; et al. COST TU1404 benchmark on macroscopic modelling of concrete
and concrete structures at early age: Proof-of-concept stage. Constr. Build. Mater. 2018, 174, 173–189.
[CrossRef]
7. Klemczak, B.; Batog, M. Heat of hydration of low-clinker cements. Pt. 1, Semi-adiabatic and isothermal tests
at different temperature. J. Therm. Anal. Calorim. 2016, 123, 1351–1360. [CrossRef]
8. Pacewska, B.; Blonkowski, G.; Wilińska, I. Investigations of the influence of different fly ashes on cement
hydration. J. Therm. Anal. Calorim. 2006, 86, 179–186. [CrossRef]
9. Klemczak, B.; Batog, M.; Pilch, M. Assessment of concrete strength development models with regard to
concretes with low clinker cements. Arch. Civ. Mech. Eng. 2016, 16, 235–247. [CrossRef]
10. Bouzoubaa, N.; Zhang, M.H.; Malhotra, V.M. Mechanical properties and durability of concrete made with
high-volume fly ash blended cements using a coarse fly ash. Cem. Concr. Res. 2001, 31, 1393–1402. [CrossRef]
11. Huang, C.H.; Lin, S.K.; Chang, C.S.; Chen, H.J. Mix proportions and mechanical properties of concrete
containing very high-volume of Class F fly ash. Constr. Build. Mater. 2013, 46, 71–78. [CrossRef]
12. Nagrockiene, D.; Pundiene, I.; Kicaite, A. The effect of cement type and plasticizer addition on concrete
properties. Constr. Build. Mater. 2013, 45, 324–331. [CrossRef]
13. PN-B-19707:2013. Cement—Special Cement—Composition, Specifications and Conformity Criteria; Polish
Committee for Standardization: Warsaw, Poland, 2013. (In Polish)
14. EN 12390-3:2009. Testing Hardened Concrete—Part 3: Compressive Strength of Test Specimens; British Standards
Institution: London, UK, 2009.
15. PN-EN 12390-5:2011. Testing Hardened Concrete—Part 5: Bending Strength of Test Specimens; British Standards
Institution: London, UK, 2011.
16. ISO 1920-10:2010. Testing of Concrete—Part 10: Determination of Static Modulus of Elasticity in Compression;
International ISO: Geneva, Switzerland, 2010.
17. PN-B-06714-23:1984. Testing—Determination of Volume Changes with Amsler Method; Polish Committee for
Standardization: Warsaw, Poland, 2013. (In Polish)
18. ACI 232.3R-14. Report on High-Volume Fly Ash Concrete for Structural Purposes; American Concrete Institute:
Farmington Hills, MI, USA, 2014.
19. Xinjie, W.; Qun, X.; Pinghua, Z. Experimental Research on the Effect of Ground Slag on Basic Tensile Creep
of Early-age Concrete. Open Construct. Build. Technol. J. 2015, 9, 68–72.
20. Japan Society of Civil Engineers. JSCE Guidelines for Concrete No. 15: Standard Specifications for concrete
structures, Design; Japan Society of Civil Engineers: Tokyo, Japan, 2011.
21. Bamforth, P.B. CIRIA C660. Early-Age Thermal Crack Control in Concrete; CIRIA, Classic House: London,
UK, 2007.
22. Klemczak, B. Prediction of Coupled Heat and Moisture Transfer in Early-Age Massive Concrete Structures.
Numer. Heat Transf. Part A Appl. 2011, 60, 212–233. [CrossRef]
23. Klemczak, B. Modelling thermal–shrinkage stresses in early age massive concrete structures—comparative
study of basic models. Arch. Civ. Mech. Eng. 2014, 14, 721–733. [CrossRef]
24. Klemczak, B.; Knoppik-Wróbel, A. Reinforced concrete tank walls and bridge abutments: Early-age
behaviour, analytic approaches and numerical models. Eng. Struct. 2015, 84, 233–251. [CrossRef]
25. Klemczak, B.; Batog, M. Heat of hydration of low clinker cements Part II—Determination of apparent
activation energy and validity of the equivalent age approach. J. Therm. Anal. Calorim. 2016, 123, 1361–1369.
[CrossRef]
26. Trinhztfy, H.W.; Jongendijk, J. Temperature development in concrete structures taking account of state
dependent properties. In Proceedings of the International RILEM Conference on Concrete of Early Ages,
Paris, France, 6–8 April 1982; pp. 211–218.
27. Neville, A.M. Properties of Concrete; Pearson Education: Harlow, UK, 2012.
Materials 2018, 11, 2207 18 of 18

28. Loredana, C.; Cuomo, M.; Greco, L. Meso-scale simulation of concrete multiaxial behaviour. Eur. J. Environ.
Civ. Eng. 2017, 21, 896–911.
29. Giorgio, I.; Scerrato, D. Multi-scale concrete model with rate-dependent internal friction. Eur. J. Environ.
Civ. Eng. 2017, 21, 821–839. [CrossRef]
30. Ramlochan, T.; Zacarias, P.; Thomas, M.D.A.; Hooton, R.D. The effect of pozzolans and slag on the expansion
of mortars cured at elevated temperature: Part I: Expansive behaviour. Cem. Concr. Res. 2003, 33, 807–814.
[CrossRef]
31. Ramlochan, T.; Zacarias, P.; Thomas, M.D.A.; Hooton, R.D. The effect of pozzolans and slag on the expansion
of mortars cured at elevated temperature: Part II: Microstructural and microchemical investigations.
Cem. Concr. Res. 2004, 34, 1341–1356. [CrossRef]
32. Altoubat, S.A.; Lange, D.A. Tensile basic creep: Measurements and behaviour at early age. ACI Mater. J.
2001, 98, 386–392.
33. Atrushi, D.S. Tensile and Compressive Creep of Early Age Concrete: Testing and Modelling. Ph.D. Thesis,
Norwegian University of Science and Technology, Trondheim, Norway, 2003.
34. Barre, F.; Bisch, P.; Chauvel, D.; Cortade, J.; Coste, J.F.; Dubois, J.P.; Erlicher, S.; Gallitre, E.; Labbé, P.;
Mazars, J.; et al. Hydration Effects of Concrete at an Early Age and the Scale Effect. In Control of Cracking in
Reinforced Concrete Structures: Research Project CEOS.fr; John Wiley & Sons: Hoboken, NJ, USA, 2016.

© 2018 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 (http://creativecommons.org/licenses/by/4.0/).