Construction and Building Materials 98 (2015) 137–145

Contents lists available at ScienceDirect

Construction and Building Materials

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

A fundamental study on compressive strength, static and dynamic

elastic moduli of young concrete
Yong Zhou a,⇑, Jie Gao a, Zhihui Sun b, Wenjun Qu a
a
Department of Structural Engineering, Tongji University, 1239 Siping Road, Shanghai 20092, PR China
b
Department of Civil and Environmental Engineering, University of Louisville, Louisville, KY 40292, USA

h i g h l i g h t s

Relationship between static and dynamic moduli of elasticity is linear.

Aggregate volume content, water-to-cement ratio and curing temperature affect E.

Aggregate volume content and maximum size are dominant factors on Ec–Ed.

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

Article history: This study investigates the influence of volume content of aggregates, maximum size and type of coarse
Received 25 April 2015 aggregates, water-to-cement ratio and curing temperature on mechanical properties, i.e. prismatic
Received in revised form 6 July 2015 compressive strength (fc), static modulus of elasticity (Ec) and dynamic modulus of elasticity (Ed) of
Accepted 12 August 2015
concrete at early age. A new equation is proposed to correlate prismatic compressive strength and elastic
Available online 24 August 2015
moduli of concrete. Based on the experimental data and the analysis results, the Ec–Ed relationship is also
proposed. It is found that the relationship between Ec and Ed is linear, and the coefficients of linear
Keywords:
relationship are analyzed by multiple regression analysis, considering aggregate content, maximum size
Early age
Impact resonance test
of the coarse aggregate, water-to-cement ratio and curing temperature. It is found that the volume
Influential factors content of aggregates is the most significant factor that influences the Ec–Ed relationship.
Multiple regression analysis Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction There are many influential factors on compressive strength and

The construction of concrete structures greatly depends on the
mechanical properties of concrete, among which compressive
strength and modulus of elasticity are mostly concerned. For
example, the proper time to remove formwork and the proper time
to apply pre-stress on concrete members are completely controlled
by these two properties. If anything improper is done before con-
crete has developed its desired properties, large deformation,
crushing of concrete or even catastrophic collapse may happen.
Therefore, knowing the early-age compressive strength and the
elastic modulus of concrete is critical to guarantee its life-time
performance.
For concrete, both the compressive strength and the elastic
modulus increase rapidly during its early age [1–3]. The design
code recommends estimating concrete’s elastic modulus based
on its 28th day’s compressive strength [4–9]. This recommended
relationship may not be applicable to concrete at early age.
⇑ Corresponding author.
E-mail address: yongzhou@tongji.edu.cn (Y. Zhou).
modulus of elasticity of concrete. Stock et al. [10] presented that
the modulus of elasticity is proportional to the volume content of
aggregate. Ranchero [11] indicated that volume content of aggre-
gates, type of coarse aggregate and water-to-cement ratio were
the most important influential factors. Johnson and Bawa [12]
found that the modulus of elasticity increases with the increase
in volume content of aggregates and decreases with the increase
in water-to-cement ratio. Yıldırım and Sengul [13] pointed out that
the modulus of elasticity could be lower if smaller aggregates were
used. All the above researches focused on concrete at the age of
28 days or older. It is not clear that how these factors influence
the modulus of elasticity at early age.
When design a concrete structure, its compressive strength and
the static Young’s modulus are used as recommended by the
design codes. However, for field measurement, concrete quality is
normally estimated via in situ non-destructive testing (NDT)
methods, among which the ultrasonic pulse velocity, the wave
reflection, and the impact echo methods are commonly used
[14–17]. These methods are dynamic methods, measuring the
dynamic modulus of concrete. Therefore, a good model to correlate

http://dx.doi.org/10.1016/j.conbuildmat.2015.08.110
0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.
138 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145

static and dynamic moduli of concrete is needed to correlate struc-

ture design and field measurement. Although the relationship
between the static and the dynamic moduli of concrete has been
suggested by some researchers [11,18–20], a model based on
design parameters, such as aggregate size and volume, water-to-
cement ratio and curing temperature is needed to close the gap
between structure design and field measurement.
This paper presents the results of a study on the development of
prismatic compressive strength and modulus of elasticity, and the
relationships between static and dynamic moduli of elasticity of
concrete within the age of 12 h to 28 days. Meanwhile, the influ-
ences of the material parameters used in designing mix propor-
tions (such as water-to-cement ratio, maximum coarse aggregate
size, coarse aggregate type, and coarse aggregate volume content,
etc.) on the relationship between static modulus (Ec) and dynamic
modulus (Ed) are investigated.

Fig. 1. Gradation curves of coarse aggregate.

2. Materials and experiments

2.1. Raw materials

Ordinary type I Portland cement was used in all the experiments. The specific
gravity of the cement was assumed to be 3.15. River sand, gravel and crushed lime-
stone were used as fine and coarse aggregates. To study the influence of maximum
coarse aggregate size on the mechanical properties of concrete, the maximum
diameters of crushed limestone were chosen as 16 mm, 20 mm and 31.5 mm,
respectively. The maximum diameter of gravel was 20 mm. The aggregate gradation
complied with GB/T14684-2011 [21] and GB/T14685-2011 [22]. In Table 1, the
cumulative particle size distributions, the specific gravity under the saturated-
surface-dry condition and the absorption capacities of the sand, gravel and crushed
limestone are listed in details. The gradation curves of all the coarse aggregates that
used in this study are plotted in Fig. 1.

Table 1
Properties of aggregates.

Crushed limestone with the diameter 5~31.5mm (C1)

Sieve size (mm) 37.5 31.5 19.0 9.5 4.75 2.36

Mass retained (%) 0 0 26 93* 100 100

3
Density (SSD) 2634 kg/m Absorption capacity 1.05%

Crushed limestone with the diameter 5~20mm (C2)

Sieve size (mm) 26.5 19.0 9.5 4.75 2.36

Mass retained (%) 0 5 72 98 100

Density (SSD) 2638 kg/m3 Absorption capacity 0.94%

Crushed limestone with the diameter 5~16mm (C3)

Sieve size (mm) 19.0 16.0 9.5 4.75 2.36

Mass retained (%) 0 9 55 96 100

3
Density (SSD) 2630 kg/m Absorption capacity 1.01%

Gravel with the diameter 5~20mm (G)

Sieve size (mm) 26.5 19.0 9.5 4.75 2.36

Mass retained (%) 0 9 66 96 100

Density (SSD) 2587 kg/m3 Absorption capacity 1.02%

River sand

Sieve size (mm) 9.50 4.75 2.36 1.18 0.6 0.3 0.15

Mass retained (%) 0 5 15 28 43 90 98

Fineness modulus 2.63- medium sand

Density (SSD) 2604 kg/m3 Absorption capacity 1.81%

*
Note: mass retained should be 90% at the sieve size 9.5 mm according to GB/T14685-2011.
 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145 139

2.2. Mixture design

In order to investigate the influence of aggregate volume content, maximum

size of coarse aggregate, aggregate type, water-to-cement ratio and curing temper-
ature on the mechanical properties of concrete, 15 mix proportions were designed
and listed in Table 2. The mass ratio of fine aggregate to total aggregate was 0.33 for
all mixture proportions. Specimen SVCA0–0.75 were designed for studying the
influence of volume content of aggregates by using the same crushed limestone
with the diameter 5–16 mm and the water-to-cement ratio of 0.5. The coarse aggre-
gates of SMSCA16–31.5 were the crushed limestone with maximum size of 16 mm,
20 mm and 31.5 mm, respectively. The coarse aggregate of SCL20 and SG20 were
the crushed limestone with the maximum diameter of 20 mm and the gravel with
the same size, respectively. SWC0.35–0.6 were designed for studying the influence
of water-to-cement ratio from 0.35 to 0.6 by using the same volume content of
aggregate (0.65). All these concrete were moisture cured under 21 °C throughout
the study. In order to study the temperature effect, for the concrete with aggregate
volume content equals 0.65, two other curing temperatures (12.5 °C and 33.5 °C)
were also used. And they are listed as the ‘‘SCT series” in Table 2. Naphthalene-
based superplasticizer was added in SVCA0.75 and SWC0.35 to achieve similar
slump to other mixtures. The mass ratio of superplasticizer to cement was 0.8%
in SVCA0.75 and 0.75% in SWC0.35.
2.3. Sample preparation
ASTM C305-06 [23] and C192-06 [24] were followed during the mixing of
SVCA0 and all others, respectively. Freshly mixed concrete was cast into
100  100  300 mm steel molds, then sealed and put into the curing chamber
within 30 min after mixing. 24 h later, the molds were removed and the specimens
were directly submerged in saturated lime water to prevent the leaching of Ca(OH)2
in concrete. Constant temperatures of 12.5 ± 1 °C, 21 ± 0.5 °C and 33.5 ± 1 °C were
used. 50 specimens were made from each mix proportion.
2.4. Test methods
The specimens of each mix proportions were tested at the age of 12 h, 1 day,
2 days, 3 days, 7 days, 14 days and 28 days. Six specimens were tested at each
age: 3 specimens for the prismatic compressive strength tests and 3 specimens
for the static and dynamic elastic modulus tests. For the moduli tests, because
the dynamic test is a NDT-test, the specimens used for this test can be used again
for the static test. Details for each test are specified as follows.
2.4.1. Impact resonance tests
Impact resonance tests were performed according to ASTM C215-08 [25], using
the longitudinal mode. Fig. 2 shows a schematic of the impact resonance test used
in this study. A miniature piezoelectric accelerometer that has a flat response from
1 to 12,000 Hz with a sensitivity of 10 mV/g at 160 Hz (LC0159 manufactured by
Lance) was mounted at the center of one end surface of the prism. A steel ball with
a diameter of 12 mm was used as the impact source and applied at the center of the
opposite end surface.
The dynamic modulus of elasticity is obtained from the following equations:
Ed ¼ DMn2 ð1Þ
D ¼ 4ðL=btÞ ð2Þ
Fig. 2. Schematic of the impact resonance test.
where Ed is the dynamic modulus of elasticity, L is the length of the specimen, b and t
are the dimensions of the prism cross section, M is the mass of the specimen, and n is
the measured fundamental longitudinal frequency.
2.4.2. Static tests
The static tests were conducted following GB/T 50081-2002 [26], using a 50 kN
universal testing machine at the age of 12 h and 1 day and a 500 kN universal test-
ing machine at the age of 2–28 days. Firstly, three replicas were used to obtain the
prismatic compressive strength, and the compressive load was applied continu-
ously at the specified loading rate. Although both cylindrical strength and prismatic
strength can be tested, the prismatic strength was chosen in this study. This is
because the Chinese code for static modulus estimation at 28 days requires the
28-day prismatic compressive strength. To keep the consistency, prisms were used
in the tests to obtain compressive strength at different ages.
After the strength tests, another three replicas (also used for dynamic tests)
were prepared for static modulus of elasticity. When measuring the static modulus,
two 100 mm linear extensometers were attached to two opposite sides of the spec-
imen to measure the axial deformation, in Fig. 3. Both extensometers were manu-
factured by Epsilon, model 3542RA2-100M-600M-ST with the measuring range of
±6 mm. The specimen was preloaded and unloaded once firstly, then loaded and
unloaded thrice to approximately 33% of the prismatic compressive strength before
loaded to failure. The stress–strain curves were plotted for each specimen. And the
slope of the stress–strain curve (up to 33% of the strength) was determined as the
static elastic modulus via linear regression.
3. Test results and analysis
3.1. Development of the mechanical properties
3.1.1. Prismatic compressive strength – fc
Due to the size effect [27,28], the prismatic compressive
strength is different from the cylindrical strength. This difference
is influenced by size, shape and placement direction of specimens
[29]. According to GB 50010-2010 [30], the ratio of the cylindrical
strength to the prismatic compressive strength is 1.18 for ordinary
concrete.
Fig. 4 plots the prismatic strength development for concrete
SMSCA20 from 12 h to 28 days. It can be seen that the compressive

Table 2
Mixture proportions.

No. Quantity (kg/m3) Mixture proportion by weight Mixture proportion by volume

Cement Water Sand Coarse aggregate
SVCA0 1211.1 605.5 0.0 0.0 1:0.5:0:0 0.38:0.61:0:0
SVCA0.4 721.7 360.9 346.0 702.5 1:0.5:0.48:0.97 0.23:0.36:0.13:0.27
SVCA0.65 415.9 208.0 562.3 1141.6 1:0.5:1.35:2.74 0.13:0.21:0.22:0.43
SVCA0.75 293.6 146.8 648.8 1317.2 1:0.5:2.21:4.49 0.09:0.15:0.25:0.50
SMSCA16 415.9 208.0 562.3 1141.6 1:0.5:1.35:2.74 0.13:0.21:0.22:0.43
SMSCA20 415.9 208.0 563.4 1143.9 1:0.5:1.35:2.75 0.13:0.21:0.22:0.43
SMSCA31.5 415.9 208.0 562.9 1142.8 1:0.5:1.35:2.75 0.13:0.21:0.22:0.43
SCL20 415.9 208.0 563.4 1143.9 1:0.5:1.35:2.75 0.13:0.21:0.22:0.43
SG20 415.9 208.0 556.1 1129.1 1:0.5:1.34:2.71 0.13:0.21:0.22:0.43
SWC0.35 509.4 178.3 562.3 1141.6 1:0.35:1.10:2.24 0.12:0.22:0.22:0.43
SWC0.5 415.9 208.0 562.3 1141.6 1:0.5:1.35:2.74 0.13:0.21:0.22:0.43
SWC0.6 370.6 222.4 562.3 1141.6 1:0.6:1.52:3.08 0.16:0.18:0.22:0.43
SCT12.5 415.9 208.0 562.3 1141.6 1:0.5:1.35:2.74 0.13:0.21:0.22:0.43
SCT21 415.9 208.0 562.3 1141.6 1:0.5:1.35:2.74 0.13:0.21:0.22:0.43
SCT33.5 415.9 208.0 562.3 1141.6 1:0.5:1.35:2.74 0.13:0.21:0.22:0.43
140 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145

Table 3
Regression coefficients of prismatic compressive strength.

Specimens f0 (MPa) A1 A2 t1 t2
SVCA0 25.702 18.521 12.103 19.416 0.881
SVCA0.4 26.662 21.346 20.703 9.981 0.402
SVCA0.65 26.166 17.855 20.929 6.328 0.583
SVCA0.75 25.625 14.676 22.671 5.712 0.746
SMSCA16 26.166 17.855 20.929 6.328 0.583
SMSCA20 25.053 14.241 19.047 6.069 0.941
SMSCA31.5 23.657 12.948 19.550 7.621 0.836
SCL20 25.053 14.241 19.047 6.069 0.941
SG20 23.307 10.075 17.560 10.681 1.787
SWC0.35 47.502 30.716 65.654 14.694 0.368
SWC0.5 26.166 17.855 20.929 6.328 0.583
SWC0.6 23.529 10.972 16.179 17.947 1.937
SCT12.5 33.132 14.743 23.896 9.985 2.689
SCT21 26.166 17.855 20.929 6.328 0.583
SCT33.5 31.513 11.376 23.512 11.178 1.618

maximum coarse aggregate size. Through this multiple regression

Fig. 3. Test setups for static modulus measurement.
analysis, it can also be seen that the water-to-cement ratio is most
important influential factor to f0, as the f0 value of SWC0.35 is twice
as much as that of SWC0.6. A1 and A2 are two coefficients that con-
trol the converging rate of the curve. They are also influenced by
water-to-cement ratio the most. t1 and t2 are two decay constants.
They are influenced by both the volume content of aggregates, the
water-to-cement ratio and curing temperature.

3.1.2. Static modulus of elasticity (Ec) and dynamic modulus of

elasticity (Ed)
Fig. 5 plots the static and the dynamic moduli for concrete
SMSCA20 from 12 h to 28 days, respectively. As shown in the fig-
ure, both moduli increase rapidly at the age of 12 h through 3 days,
while they increase slowly at the age of 3–28 days. Both Ec and Ed
at the age of 3 days reach 80% of the values of 28 days. Meanwhile,
the increasing rates of both Ec and Ed are faster than that of fc dur-
ing the age of 12 h to 3 days. Similar results were found by Lew and
Reichard [31]. All other specimens used in this study showed the
same trend. Therefore an exponential equation is proposed to rep-
resent the development tendency of Ec an Ed with age:
Fig. 4. Development of prismatic compressive strength of SMSCA20.
Ec ðtÞ or Ed ðtÞ ¼ E0 þ B1 et=g1 þ B2 et=g2 ð4Þ

where t represents the hydration age (in days), E0, B1, B2, g1 and g2
strength increases rapidly at the age of 12 h to 7 days, while it
are coefficients.
increases slowly at the age of 7–28 days. And the compressive
strength at the age of 7 days was approximately 80% of that at
28 days. All other specimens used in this study showed the same
characteristics. To further study the development of strength
growth, an exponential equation shown as Eq. (3) was proposed
to simulate the development trend.

f c ðtÞ ¼ f 0 þ A1 et=t1 þ A2 et=t2 ð3Þ

where t represents the hydration age (in days), f0, A1, A2, t1 and t2
are coefficients from regression.
It is clearly seen that Eq. (3) can depict the development ten-
dency of prismatic compressive strength, and the fitted result of
Adj.R2 is close to 1. For all the tested concrete, the regression coef-
ficients f0, A1, A2, t1 and t2 are listed in the following Table 3.
From both Eq. (3) and the Table 3, it can be seen that the coef-
ficient f0 represents the ultimate strength of the concrete (when
t ? 1, which has a value pretty close to the 28 days strength. By
comparing specimens SVCA0 through SVCA0.75, it can be seen that
changing the volume content of aggregates does not affect the ulti-
mate strength gains of concrete. The f0 values of SMSCA16 through
SMSCA31.5 indicate a decrease of the strength with an increase of Fig. 5. Development of the static and the dynamic modulus for SMSCA20.
 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145 141

Apparently, the curve of Eq. (4) is closer to the experimental Table 6

data, and the fitted result of Adj.R2 is close to 1. Therefore, Eq. (4) Estimating equations in different codes.

can depict the development tendency of the modulus of elasticity. Code Estimating equation Units
Similar to the compressive strength, the E0 in the equation indicate pffiffiffiffiffi
ACI 318-08 Ec ¼ 4700 f c fc: MPa, Ec: MPa
the ultimate modulus of concrete (when t ? 1; B1 and B2 are two pffiffiffiffiffi
ACI 363-08 Ec ¼ 3300 f c þ 6900 fc: MPa, Ec: MPa
shape factors that control the converge rate of the curve; and g1 pffiffiffiffiffi
NZS 3101-2006 Ec ¼ 3320 f c þ 6900 fc: MPa, Ec: MPa
and g2 are two decay constants. pffiffiffiffiffi
CSA A23.3-04 Ec ¼ 4500 f c fc: MPa, Ec: MPa
The curve fitting results are listed in Tables 4 and 5 for all the EC-2 Ec = 22(fc/10)0.3 fc : MPa, Ec : GPa
coefficients. From the tables, it can be seen that E0 has a value GB 50010 2
Ec ¼ 2:2þ34:7=f
10 fc: MPa, Ec: MPa
c
pretty close to the 28 days modulus of elasticity. To be different
from the compressive strength, the volume content of aggregates
is the most significant influencing factor on the E0 values. B1 and
B2 are also affected by aggregate volume content the most. g1
and g2 are more sensitive to the water-to-cement ratio and the cur-
ing temperature.

3.2. Ec and Ed vs. fc

The relationship between Ec and fc has been recommended by

design codes of many countries [4–9], shown in Table 6. Fig. 6 com-
pares the values estimated by the equations in different codes with
the experimental data of SWC0.5. The calculated values of ACI 318
and CSA A23.3 at the age of 12 h, and EC-2 at the age of 1 day, are
closest to the experimental data, respectively. After the age of
2 days, all the calculated values are smaller than the experimental
data.

Table 4
Regression coefficients of the static modulus of elasticity.

Specimens E0 (GPa) B1 B2 g1 g2
Fig. 6. Comparison of the calculated values and the experimental data of SWC0.5.
SVCA0 12.35 5.57 14.77 10.17 0.56
SVCA0.4 25.11 15.53 68.66 4.64 0.26
SVCA0.65 35.91 12.69 122.14 6.41 0.27
SVCA0.75 38.80 11.47 110.87 4.14 0.34 All other mix proportions have similar characteristics.
Therefore, Ec would be underestimated by using the abovemen-
SMSCA16 35.91 12.69 122.14 6.41 0.27
SMSCA20 37.89 10.00 48.84 14.43 0.69 tioned codes. Previous research found that during its early age,
SMSCA31.5 35.29 10.60 61.09 4.03 0.49 concrete compressive strength is mainly governed by the strength
SCL20 37.89 10.00 48.84 14.43 0.69 of its paste matrix, the flaw size, and the ITZ properties. However,
SG20 32.43 12.32 70.97 5.38 0.39 the elastic modulus of concrete is more influenced by its aggregate
SWC0.35 40.89 9.60 128.32 7.78 0.32 contents and properties [32]. This hints that using one-fit-all equa-
SWC0.5 35.91 12.69 122.14 6.41 0.27 tions to correlate compressive strength and elastic modulus will
SWC0.6 37.92 14.03 48.29 18.74 0.65 not lead to reliable results. Influencing factors, such as aggregate
SCT12.5 38.78 9.66 45.62 17.63 1.16 volume fraction, maximum aggregates size (dominant influence
SCT21 35.91 12.69 122.14 6.41 0.27 on ITZ), water-to-cement ratio, and hydration age, etc. should be
SCT33.5 37.19 10.40 39.98 5.63 0.51
included.
Different from Venkiteela et al. [32], the relationship between
Ed and fc was studied. The equations in Table 6 can be modified
to accommodate the dynamic modulus of elasticity because of
Table 5
the linear relationship between Ec and Ed (in GPa). The equations
Regression coefficients of the dynamic modulus of elasticity.
can be rewritten as follows:
Specimens E0 (GPa) B1 B2 g1 g2
n
SVCA0 13.53 6.11 15.76 7.74 0.56 Ec or Ed ¼ af c þ b ð5Þ
SVCA0.4 28.55 15.51 64.76 4.71 0.30
SVCA0.65 41.93 13.39 99.86 6.37 0.34 where a, b and n are coefficients.
SVCA0.75 45.01 15.04 177.47 3.29 0.26 Fig. 7 shows the result of Eq. (5) and experimental data of
SMSCA16 41.93 13.39 99.86 6.37 0.34 SWC0.5. The regression data are very close to experimental data,
SMSCA20 40.82 10.80 65.68 6.44 0.51 and both R2 values are so close to 1. Coefficient n is nearly 0.295,
SMSCA31.5 40.80 16.42 82.23 2.91 0.36 and is not subjected to the influential factors. Coefficient a governs
SCL20 40.82 10.80 65.68 6.44 0.51 the shape and converging rate of the relationship. As shown in
SG20 39.09 16.40 81.50 4.37 0.35 Table 7, a is more related to the volume content of aggregate
SWC0.35 46.89 9.86 130.75 7.68 0.33 (Vagg) and maximum coarse aggregate size (Magg). Coefficient b is
SWC0.5 41.93 13.39 99.86 6.37 0.34 influenced by the volume content of aggregate, maximum coarse
SWC0.6 41.01 14.35 66.98 9.62 0.48
aggregate size and water-to-cement ratio (w/c). Both this research
SCT12.5 41.94 6.47 48.79 8.91 1.33 and previous research [32] show that curing temperature does not
SCT21 41.93 13.39 99.86 6.37 0.34
affect fc–Ec (or fc–Ed) relationships. The detailed relationships are
SCT33.5 43.93 8.92 43.02 13.94 0.62
expressed as follows:
142 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145

Fig. 7. Relationship between Ec or Ed and fc of SWC0.5. Fig. 8. Relationship between Ec and Ed of SMSCA31.5.

Table 7
Coefficients a and b for each specimen. Table 8
Coefficient c and d for the relationship between Ec and Ed.
Specimens Ec (GPa) and fc (MPa) Ed (GPa) and fc (MPa)
Specimens c d
a b n a b n
SVCA0 0.894 0.023
SVCA0 7.00 5.63 0.302 7.87 6.07 0.301
SVCA0.4 0.885 0.089
SVCA0.4 12.89 9.21 0.301 14.72 9.56 0.299
SVCA0.65 0.887 0.134
SVCA0.65 18.51 12.03 0.292 20.02 11.17 0.300
SVCA0.75 0.951 0.401
SVCA0.75 20.27 13.49 0.294 22.66 11.27 0.282
SMSCA16 0.887 0.260
SMSCA16 18.51 12.03 0.292 20.02 11.17 0.300
SMSCA20 0.917 0.211
SMSCA20 18.62 13.40 0.298 20.32 12.34 0.298
SMSCA31.5 0.921 0.238
SMSCA31.5 19.54 14.03 0.299 21.56 13.00 0.296
SCL20 0.917 0.211
SCL20 18.62 13.40 0.298 20.32 12.34 0.298
SG20 0.882 0.229
SG20 18.07 11.51 0.286 20.19 10.16 0.289
SWC0.35 0.951 0.390
SWC0.35 18.25 12.49 0.293 20.03 9.86 0.285
SWC0.5 0.887 0.134
SWC0.5 18.51 12.03 0.292 20.02 11.17 0.300
SWC0.6 0.901 0.230
SWC0.6 18.42 12.04 0.297 20.41 10.76 0.297
SCT12.5 0.914 0.243
SCT12.5 18.16 12.22 0.292 20.19 11.02 0.280
SCT21 0.887 0.134
SCT21 18.51 12.03 0.292 20.02 11.17 0.300
SCT33.5 0.971 0.353
SCT33.5 18.11 12.39 0.298 20.01 10.65 0.286

a ¼ 17:88V agg þ 0:08M agg þ 5:27 ð6Þ

b ¼ 7:41V agg  4:23V 2agg  0:12Magg þ 1:79ðw=cÞ  4:60
3.3. Relationship between Ec and Ed
The study shows that the dynamic modulus of elasticity is
always greater than the static modulus of elasticity at each specific
age. Taking SMSCA31.5 as an example, a linear relation between Ec
and Ed can be observed as shown in Fig. 8. This relationship can be
described by Eq. (8):
Ec ¼ cEd þ d
where c and d are regression coefficients.
For all the used concrete, this relationship fit. The values of coef-
ficients c and d are listed in Table 8. From the Table 8, it can be seen
that the c and d values are certainly influenced by the volume con-
tent of aggregate, maximum coarse aggregate size, water-to-
cement ratio and curing temperature. The details are discussed in
the following sections.
3.4. Influential factors on Ec–Ed relationship
Fig. 9(a) shows the influence of volume content of aggregate on
elastic moduli. Because aggregate is generally more rigid than fresh
paste in ordinary concrete, increasing the volume content of aggre-
ð7Þ gate will lead to the increase of both Ec and Ed. The elastic modulus
of SVCA0.65 is far greater than SVCA0 and SVCA0.4. This indicates
that the volume content of aggregates is the most important influ-
ential factors on Ec and Ed. From the Table 8 and Fig. 9(b), it can be
seen that the value of c increased significantly when the volume
concrete of aggregates was increased from 65% to 75%. The d values
keep decreasing with the increase of aggregate contents. This indi-
cates that compared to Ec, Ed is more sensitive to the change in
aggregate volume content.
Fig. 10 plots the influence of maximum coarse aggregate size on
ð8Þ Ec, Ed, and Ec–Ed relationship. As shown in Fig. 10(a), it seems that
reducing the maximum aggregate size from 31.5 mm to 20 mm or
16 mm do not substantially affect the Ec and Ed. In Fig. 10(b), the
experimental curves of SMSCA16, SMSCA20 and SMSCA31.5 are
almost overlapped. From Table 8, it can be seen that the slope (c
value) increases with the increase of aggregate size, however, the
intercept (d value) does not change much with the aggregate size.
Therefore, the maximum aggregate size would only affect the slope
of the Ec–Ed relationship.
Fig. 11 shows the influence of coarse aggregate type on Ec, Ed,
and Ec–Ed relationship. Ec and Ed of the specimens with crushed
limestone are greater than those with gravel. The reason could
be that the used limestone has relatively higher modulus than that
of the gravel. It could also due to better ITZ in SCL 20 specimens,
 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145 143

Fig. 9. Influence of volume content of aggregate.

Fig. 10. Influence of maximum size of coarse aggregate.

because the surface of crushed limestone is rougher than that of
gravel, so a better adhesion between crushed limestone and
cement paste is ensured. However, the aggregate type does not
affect the relationship between Ec and Ed. One can notice that the
two lines in Fig. 11(b) are almost overlapped. This can also be
reflected by the similar values of c and d (for SLG20 and SG20) in
Table 8.
Fig. 12 shows the influence of water-to-cement ratio on Ec, Ed,
and Ec–Ed relationship. From Fig. 12(a), it can be seen that for a
given age, both Ec and Ed decrease with the increase of the
water-to-cement ratio within the studied range. A lower water-
to-cement ratio will lead to a stronger paste matrix, which
enhances the elastic moduli of concrete. As shown in Fig. 12(b)
and Table 8, the values c and d are not influenced by the water-
to-cement ratio. This hints that the aggregates are the dominant
phase that governs the Ec–Ed relationship.
The temperature influences on Ec, Ed, and Ec–Ed relationship are
given in Fig. 13. In Fig. 13(a), it shows that for a lower curing
temperature (e.g. 12.5 °C) both Ec and Ed would approach their ulti-
mate values much slower than those cured under higher tempera-
tures. For the specimens cured in 33.5 °C, higher increase rates of Ec
and Ed can be expected during its early age that leads to a quick
converge to their final values. SCT21 represents the specimens
cured in normal temperature (21 °C). One should notice that in
Table 8, although the values c and d varied a lot, a direct correlation
between temperature and c, d values cannot be found. This again
indicates that aggregate is the dominant phase that affect the Ec–
Ed relationship, as temperature mainly influence the hydration of
the cement paste.
According to the above analysis, the linear coefficients c and d in
Eq. (8) are influenced by volume content of aggregate and the max-
imum aggregate size. Multiple regression analysis was applied to
establish the relationship between the linear coefficients c and d
and the above influential factors. They are listed as follows:
c ¼ 0:048V agg þ 0:0005Magg þ 0:8755 ð9Þ
144 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145

Fig. 11. Influence of type of coarse aggregate.

Fig. 12. Influence of water-to-cement ratio.

Fig. 13. Influence of curing temperature.

 Y. Zhou et al. / Construction and Building Materials 98 (2015) 137–145
d ¼ 0:855V 3agg  0:023 ð10Þ
where Vagg is volume content of aggregate, ranges from 0 to 0.75,
Magg is maximum size of coarse aggregate, ranges from 16 mm to
31.5 mm.
4. Conclusion
The study confirms that prismatic compressive strength, static
modulus of elasticity and dynamic modulus of elasticity are influ-
enced by volume content of aggregate, maximum size and type of
coarse aggregate, water-to-cement ratio and curing temperature.
The following conclusions can be drawn:
(1) Prismatic compressive strength, static modulus of elasticity
and dynamic modulus of elasticity increase rapidly at very
early age, and then increase slowly. The development
tendency with age can be fit with an exponential decay
equation.
(2) The relationship between static (or dynamic) modulus of
elasticity and prismatic compressive strength is recom-
mended by a polynomial equation.
(3) Aggregate volume content, water-to-cement ratio and
curing temperature affect the development of both the static
and the dynamic moduli of elasticity. However, water-
to-cement ratio and curing temperature does not affect the
correlation between Ed and Ec.
(4) A linear relationship between static modulus Ec and the
dynamic modulus Ed can be found. The relationship is
mainly governed by the aggregate phase in concrete.
Aggregate volume content and maximum size are the two
dominant factors that govern the Ec–Ed correlation.
It should be noted that the above mentioned results are only
applicable to the given conventional concrete. Other types of con-
crete may have different relations and parameters of correlation.
Acknowledgments
This study was financially supported by National Natural
Science Foundation of China (Grant No. 51208373) and the
Natural Science Foundation of Shanghai, China (Grant No.
14ZR1443300). The support from the Civil and Environmental
Engineering Department, University of Louisville, is also
appreciated.
References
[1] M.a.a Abd elaty, Compressive strength prediction of Portland cement concrete
with age using a new model, HBRC J. 10 (2) (2014) 145–155.
[2] J.K. Kim, Y.H. Moon, S.H. Eo, Compressive strength development of concrete
with different curing time and temperature, Cem. Concr. Res. 28 (12) (1998)
1761–1773.
[3] R. Madandoust, J.H. Bungey, R. Ghayidel, Prediction of the concrete
compressive strength by means of core testing using GMDH-type neural
network and ANFIS models, Comput. Mater. Sci. 51 (1) (2012) 261–272.
145
[4] American Concrete Institute, ACI 318-08 Building Code for Structural Concrete,
Detroit, USA, 2008.
[5] American Concrete Institute, ACI 363R-08 State of the Art Report on High
Strength Concrete, Detroit, USA, 2008.
[6] New Zealand Standard, Concrete Structures Standard, NZS 3101:2006, The
Design of Concrete Structures, Wellington, New Zealand, 2006.
[7] CSA Technical Committee, Reinforced Concrete Design, A23.3-04, Design of
Concrete Structures, Rexdale, Ontario, Canada, 2004.
[8] European Committee for Standardization, Eurocode 2: Design of Concrete
Structures, Brussels, Belgium, 2004.
[9] Code For Design of Concrete Structures (GB 50010-2010), State Development
Planning Commission of the People’s Republic of China, Beijing.
[10] A.F. Stock, D.J. Hannantt, R.I.T. Williams, The effect of aggregate concentration
upon the strength and modulus of elasticity of concrete, Mag. Concr. Res. 31
(109) (1979) 225–234.
[11] J.L. Ranchero, Analyzing and determining relationships in elastic properties of
concrete using wave propagation and vibration and uniaxial compression (a
thesis submitted in partial fulfillment of the requirements for the degree of
Master of Science in Civil Engineering), University of Illinois at Urbana-
Champaign, Urbana, Illinois, 2005.
[12] J.W.S. de Graft-Johnson, N.S. Bawa, Effect of mix proportion, water-cement
ratio, age and curing conditions on the dynamic modulus of elasticity of
concrete, Build. Sci. 3 (3) (1969) 171–177.
[13] H. Yıldırım, O. Sengul, Modulus of elasticity of substandard and normal
concretes, Constr. Build. Mater. 25 (4) (2011) 1645–1652.
[14] G. Trtnik, F. Kavcic, G. Turk, Prediction of concrete strength using ultrasonic
pulse velocity and artificial neural networks, Ultrasonics 49 (1) (2009) 53–60.
[15] S.V. Kolluru, J.S. Popovics, S.P. Shah, Determining elastic properties of concrete
using vibrational resonance frequencies of standard test cylinders, Cem. Concr.
Aggr. 22 (2) (2000) 81–89.
[16] Z. Sun, T. Voigt, S.P. Shah, Temperature effects on strength evaluation of
cement-based materials with ultrasonic wave reflection technique, ACI Mater.
J. 102 (4) (2005) 272–278.
[17] T. Voigt, Z. Sun, S.P. Shah, Comparison of ultrasonic wave reflection method
and maturity method in evaluating early-age compressive strength of mortar,
Cem. Concr. Compos. 28 (4) (2006) 307–316.
[18] A.M. Neville, Properties of Concrete, Pitman Books Limited, New York, 1981.
[19] S.H. Han, J.K. Kim, Effect of temperature and age on the relationship between
dynamic and static elastic modulus of concrete, Cem. Concr. Res. 34 (7) (2004)
1219–1227.
[20] F.D. Lydon, R.V. Balendran, Some observations on elastic properties of plain
concrete, Cem. Concr. Res. 16 (3) (1986) 314–324.
[21] Sand for Construction (GB/T14684-2011), China Building Materials Federation,
Beijing.
[22] Pebble and Crushed Stone for Construction (GB/T14685-2011), China Building
Materials Federation, Beijing.
[23] American Society for Testing Material, ASTM C305-06 Standard Practice for
Mechanical Mixing of Hydraulic Cement Pastes and Mortars of Plastic
Consistency, West Conshohocken, Pennsylvania, USA, 2006.
[24] American Society for Testing Material, ASTM C192-06 Standard Practice for
Making and Curing Concrete Test Specimens in the Laboratory, West
Conshohocken, Pennsylvania, USA, 2006.
[25] American Society for Testing Material, ASTM C215-08 Standard Test Method
for Fundamental Transverse, Longitudinal, and Torsional Resonant Frequencies
of Concrete Specimens, West Conshohocken, Pennsylvania, USA, 2008.
[26] Standard for Test Method of Mechanical Properties on Ordinary Concrete (GB/
T 50081-2002), China Academy of Building Research, Beijing.
[27] Z.P. Bazant, Y.J. Xiang, Size effect in compression fracture: splitting crack band
propagation, J. Eng. Mech. ASCE 123 (2) (1997) 162–172.
[28] J.K. Kim, S.T. Yi, Application of size effect to compressive strength of concrete
members, Sadhana Acad. Proc. Eng. Sci. 27 (2002) 467–484.
[29] S.T. Yi, E.K. Yang, J.C. Choi, Effect of specimen sizes, specimen shapes, and
placement directions on compressive strength of concrete, Nucl. Eng. Des. 236
(2) (2006) 115–127.
[30] Code for Design of Concrete Structures (GB 50010-2010), China Academy of
Building Research, Beijing.
[31] H.S. Lew, T.W. Reichard, Mechanical properties of concrete at early ages, ACI J.
Proc. 75 (10) (1978) 533–542.
[32] G. Venkiteela, Z. Sun, H. Najm, Prediction of early age normal concrete
compressive strength based on dynamic shear modulus measurements, J.
Mater. Civ. Eng. 25 (1) (2013) 30–38.