Construction and Building Materials 105 (2016) 189–195

Contents lists available at ScienceDirect

Construction and Building Materials

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

A strength developing model of concrete under sustained loads

Bing Han a,b,⇑, Ming-Hui Liu a, Hui-Bing Xie a, Yu-Ping Liu c
a
School of Civil Engineering, Beijing Jiaotong University, Beijing, PR China
b
Key Laboratory of Transportation Infrastructure Safety and Risk Management, Ministry of Transportation, Beijing, PR China
c
Department of Civil and Environmental Engineering, University of Houston, Houston, TX, USA

h i g h l i g h t s

Maximum past thickness in concrete was modified considering the effect of sustained load.

Strength developing rules of concrete under sustained load were explored based on compressible packing model.

Experiments were executed as a case study to suggest an available process to determine the parameters in the model.

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

Article history: The strength of concrete will be developed as the age increases. When a sustained load is applied on the
Received 17 July 2015 concrete, the development rules will be changed. Research on these rules is needed to estimate more
Received in revised form 11 November 2015 rationally the developing strength of concrete used in structures than those without considering the
Accepted 9 December 2015
action of sustained loads. In this paper, the development rule of compressive strength of concrete under
Available online 17 December 2015
sustained loads is explored. According to the mechanism assumption, a new model based on the
Compressible Packing Model (CPM) for concrete compressive strength is proposed. For calibrating the
Keywords:
model, experiments are carried out. Compressive strength of concrete were determined at four different
Concrete
Sustained load
ages and were compared with those of specimens without any sustained loads applied. Comparison of
Compressive strength the results by the model and by the experiments show the model presented in this paper can be suit
Compressible packing model for concrete strength development of concrete under sustained loading.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction
It is common sense that concrete structures have to bear
sustained loads in their service stage, especially concrete in
prestressed structures. As concrete ages, its strength will change.
In fact, most of reported researches on the development of con-
crete strength were based on the condition without sustained load
[1]. Although there are some studies that explore the aging charac-
teristics of the concrete material itself without consideration of
sustained load, the strength of concrete in a structure, which are
very important to evaluate the true behavior of concrete struc-
tures, should be studied on the basis of sustained loading
conditions.
Sustained loading may change the microstructure of concrete,
resulting in the change of the material performance. For example,
creep takes place under long-term loading because of the viscous
⇑ Corresponding author at: School of Civil Engineering, Beijing Jiaotong
University, Beijing, PR China.
E-mail address: bhan@bjtu.edu.cn (B. Han).
flow of cement gel and the development of internal micro cracks.
Some researchers found sustained loading at high stress levels
(lower than the compressive strength) may lead to concrete
destruction, and the higher stress level is, the shorter time is
needed to failure [2]. Because of this, engineers thought that sus-
tained loads had negative effects on concrete strength, and con-
crete strength was underestimated in practice. However, during
the service stage, concrete usually works in its elastic state, which
corresponds to stress lower than about 40% of the material com-
pressive strength. Such a low stress is not enough to induce unsta-
ble damage, which will decrease strength in concrete, and
reversely, the concrete strength increases [3–7]. Nevertheless, all
of these studies have not given a clear mechanism on the strength-
ening effect of sustained loading on mechanical properties.
In fact, concrete is a kind of mixture material with complicated
mechanical performance and has significant time-dependent prop-
erties. To obtain a more rational description of concrete strength
development, lots of research on modeling has been done [8,9].
Mikulic et al. [10] classified all these models into three categories,
i.e., (1) models that predict the concrete compression strength

http://dx.doi.org/10.1016/j.conbuildmat.2015.12.067
0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.
190 B. Han et al. / Construction and Building Materials 105 (2016) 189–195
based on its component; (2) models based on homogenous
assumptions, such as models based on maturity methods [11–14]
and the degree of hydration concept [15]; and (3) models based
on composition and the age of concrete, such as the Exponential
Cement Model proposed by Popovics [16] in 1998 and the
Compressible Packing Model (CPM) proposed by De Larrard [17]
in 1999. Most of these models are semi-empirical and semi-
theoretical.
2. Modeling strength development of loaded concrete
Compressible Packing Model (CPM) proposed by De Larrard [17]
has a clear description concerning the composition of the concrete
material and can be conveniently used to develop a model of
concrete for a specific problem. This paper proposes a model for
strength development of concrete under sustained loads on the
basis of CPM.
2.1. Strength development model based on CPM
Assuming concrete is a kind of binary mixture containing heavy,
hard and inert aggregates spread randomly in a cement matrix, it is
considered to be somewhat homogeneous. In such a composite
structure, the weakest position is located within the cement
matrix, which should be the primary focus of attention. A strength
development model for a hardened cement matrix is constructed
by taking into account the effects of aggregate properties and the
topological structure of the aggregates.
Hardened cement matrix is regarded as a kind of porous mate-
rial whose volume fraction will have an effect on compressive
strength [18]. De Larrard [17] found that there is more un-
hydrated content in cement matrix with low water cement ratio,
and suggested a 28-day compressive strength model as follows
[17]:

2:85
Vc
f m ¼ 11:4Rc28
Vc þ Vw þ Va
where fm is the 28-day compressive strength of cement matrix cur-
ing hermetically (MPa); Rc28 is 28-day standard strength of cement
(MPa); Vc, Vw and Va stands for volume of cement, water and aggre-
gate, respectively (mm3).
A unified physical parameter named maximum paste thickness
(MPT) was proposed by De Larrard to take account for the
maximum size and volume fraction of aggregates. Image in a dry
accumulation of aggregates, the maximum stress should occur at
junction points between aggregates [19]. If cement matrix is
injected into the accumulation, the aggregates should spread
evenly, and a layer of cement matrix would form between two
adjacent aggregates. The cement matrix layer is under high stress,
i.e. MPT. The forming process of the MPT is shown as Fig. 1 [17].
MPT is written as

rffiffiffiffiffi  Supposing the values of external work are the same regardless
g whether or not load is applied, that is W = W0 . The equivalence
MPT ¼ D 3 1
g relation can be expressed as
where D is the maximum size of the aggregate (mm); g is volume of
aggregate in unit volume after spreading (mm3); g* is the aggregate
volume before the cement matrix injected, also called stacking
density (mm3).
Concrete strength development can be expressed by [17]
8 stant related to concrete properties (MPa); a = N/(fcAc); N refers to
  2:85 
>
> external force (N); fc refers to concrete compression strength at a
>
>
>
f cm ðtÞ ¼ 13:4R c28 A lgðt=28Þ þ 1 þ q wþa
c ceq ðtÞ MPT 0:13
< h   i
>
> ceq ðtÞ ¼ c 1 þ 1:1 1  exp ðB lg t þ PÞ fca
>
> Substitute Eqs. (6) into (3), and the physical effect of load on
>
: f ðtÞ ¼ pfcm ðtÞ
c qfcm ðtÞþ1
where qc is cement density (kg/m3); t stands for time (d); A is con-
stant related to cement properties; B and P are constant related to
fly-ash properties; p and q is constant related to aggregate type;
ceq is equivalent cement mass in concrete at a given age
ceq ¼ c½1 þ Wðf a =cÞ ð4Þ
where c refers to cement mass (kg); fa stands for fly-ash mass in unit
volume of concrete (kg); W can be written as seen in Eq. (5)


fa
W ¼ Wmax 1  exp K p ð5Þ
c
where Kp is activity coefficient reflecting strength contribution of
fly-ash, and linear relation with lg t; Wmax is a coefficient related
to maximum mass of lime precipitation, and optimal value is sug-
gested as 1.10.
2.2. Modified CPM considering sustained load
2.2.1. Modification for physical effects of sustained loads
On the basis of the analysis above, concrete compaction caused
by sustained load mainly occurs in the cement matrix. As a result,
the volume of the cement matrix is reduced, and the aggregate vol-
ume can be considered unchanged based on the assumption that
aggregates are incompressible. Thus the aggregate content in a unit
concrete volume, g, will increase, inducing an MPT decrease.
According to CPM, concrete can be seen as a stressed accumula-
tion of dry aggregates injected into cement matrix in batches [17].
The initial stress in the pressure system is called r0, as shown in
Fig. 2. It should be a constant depending on the characteristics of
concrete itself and may also be related to the internal energy of
concrete at the loading age, or the strength of concrete or topolog-
ical properties between different components. When the accumu-
lation has no load action, the volume expands uniformly along the
grid posed by the center of gravity of the aggregates during the
process of injecting cement matrix, and MPT is formed at the junc-
ð1Þ
tion point of two aggregates. The expansion coefficient is set as k.
During the expansion, outside work acts on the system overcoming
r0, and the external work is set as W. The expansion coefficient
becomes k0 after the injection of cement matrix when stress is
equal to afc0, and the maximum thickness of paste changes to MPT0 ,
where k0 < k and MPT0 < MPT. During the process, outside work, W0 ,
acts on the system to overcome r0 + afc0. In Fig. 2, the comparison
of MPT formations in the two situations is shown.
Now take a coarse aggregate in the MPT as an example based on
the spherical model in which aggregates are supposed to be in a
shape of sphere. In an unstressed state, force on the coarse
aggregates is r0pd2/4 and the external forces do negative work
to result in a displacement of MPT. In a stressed state, force on
the coarse aggregates is (r0 + afc0)pd2/4 and the negative work
leads a displacement of MPT0 .
ð2Þ
r0 p4d  MPT ¼ ðr0 þ af c0 Þ p4d  MPT 0
2 2
ð6Þ
) MPT 0 ¼ r0 þra0 f MPT
c0
where r0 stands for initial stress in the accumulation and is a con-
given age (MPa); Ac refers to cross-section area of concrete member
ð3Þ (mm2); fc0 is strength of concrete at the time of loading (MPa).
strength development of concrete can be reflected by stress factor
[r0/(r0 + afc0)]0.13.
 B. Han et al. / Construction and Building Materials 105 (2016) 189–195 191

Fig. 1. Formation process of MPT.

Fig. 2. Comparison of MPT formation progress under different stress conditions.

2.2.2. Modification for chemical effects of sustained loads degree of hydration of cement, cM, will be introduced into the
Since the extra hydration reaction caused by stress will develop impact factor of time of strength of concrete in the absence of
along with time, an impact factor of the extra hydration reflecting forces. Then
192 B. Han et al. / Construction and Building Materials 105 (2016) 189–195


t naphthalene-based superplasticizer were used to produce con-
dðtÞ ¼ cM þ A lg ð7Þ
28 crete. Fly ash was used to substitute 20% content of the cement.
Mix proportion is listed in Table 1.
where c is a parameter corresponding to degree of hydration and
related to properties of concrete and stress; M stands for degree 3.2. Specimen design
of hydration.
A model for calculating M is introduced here [15] To achieve the experimental objective, two types of specimens
8 kðtt 0 Þ were designed, namely loaded specimens for compression
>
> M ¼ ff28c ¼ b 1þkðtt
>
>
0
Þ
strength, un-loaded specimens for compression strength. The spec-
<
k ¼ A0 exp E imens for compression strength were 100  100  100 mm cubes.
RT ð8Þ
>
>
> b ¼ 1 þ ð28t0 Þk
1
All of loaded specimens were imposed with a load at an age of
>
: 7 days, and were tested for their compressive strength at the ages
of 14, 28, and 60 days. At the same time, the reference un-loaded
where f28 is concrete compression strength at age of 28 d (MPa); t0 specimens were tested after curing at 7, 14, 28, and 60 days. Totally
is the time when concrete strength starts to develop, t0 = 0 if t  t0 there are 7 groups of specimens with 3 specimens in each group.
(d); k is a rate constant related to cement hydration reaction; A0 is a The amount of the load imposed on the specimens was equal to
frequency factor related to intermolecular collisions during the 20% of the 7-day compressive strength of standard cured concrete.
chemical reaction and molecular activity involving in the hydration, The quantity and numbering of specimens are listed in Table 2.
A0 = e15.46 = 5,178,365; E is surface activation energy with a unit of Concrete cube compressive strength at 7 days was measured as
J/mol; R is the universal gas constant with a value of 8.341 J/mol K, 41.6 MPa, resulting in a sustained load of 8.32 MPa. Concrete prism
and E/R = 4978 K; T is curing temperature in unit of K; b is ratio of compressive strength at 7 days in the test was 32.5 MPa, so the
ultimate strength and 28-day strength, here t0 = 0. sustained load on the prism was 6.5 MPa.

2.2.3. Model establishment 3.3. Testing procedure

Taking above consideration of physical and chemical effects
into Eq. (3), compression strength development of concrete under In order to ensure applied loads were effective, special loading
sustained loads can be expressed as follows equipment was used as shown in Fig. 3. The device was designed

8  2:85  
>
 kðtt0 Þ  t
 
> f
> cm ðtÞ ¼ 13:4R c 1 þ 1
 þ A lg þ 1 þ q wþa
 r0 þra0 f  MPT 0:13
>
>
c28 28k 1þkðtt0 Þ 28 c ceq ðtÞ c0
>
>

<
k ¼ A0 exp E
h RT   i ð9Þ
>
>
>
> ceq ðtÞ ¼ c 1 þ 1:1 1  exp ðB lg t þ PÞ fca
>
>
>
: ðtÞ
f c ðtÞ ¼ qfpfcmcmðtÞþ1

in which, c, r0, p, q, A, B, and P is undetermined parameters, and according to the post-tensioned principle, i.e. a load was applied
can be obtained by regression according to experimental data. by a jack. When the load achieved the requirement, nuts on the
top of steel rods were tightened, and the concrete specimens were
pressured. For the case with unloading, the stress was adjusted
3. Experimental program
according to a pressure sensor and foil gauge adhered on the steel
rods. Three specimens for compressive strength can be loaded at
Experiments were designed to verify the effect of sustained load
the same time. Concrete blocks were placed between specimens
on the compression strength and provide data for regression of
and steel plates to avoid uneven stress.
coefficients in Eq. (9), referenced specimens without sustained
An important issue in loading is to avoid eccentricity. Speci-
loads applied were employed.
mens were laid on the horizontal bottom plate. It was necessary
that the bearing surface was parallel to the top surface when the
3.1. Mix design specimen was in the mold, and the centers of specimen, concrete
blocks, jack, and pressure sensor were made to coincide in a line
In the experiments, materials including ordinary Portland as accurately as possible. The middle plate was placed horizontally
cement (with 54.9 MPa of the 28-day standard compressive
strength), fly ash (classified as I in Chinese standard [20]), contin-
Table 2
uous grading macadam with diameters from 5 to 25 mm, medium
Quantity and numbering of specimens.
sand with diameters from 0.08 to 2.0 mm, as well as a kind of
Items of testing Stress level (%) Age Quantity
7d 14 d 28 d 60 d
Table 1
Mix proportion of concrete (kg/m3). Serial no.
fc 0 F0-I F0-II F0-III F0-IV 4  3 = 12
Cement Fly ash Sand Macadam Water Water reducing agent
20 – F2-II F2-III F2-IV 33=9
336 84 809 1029 162.5 2.4
Note: fc refer to concrete compressive strength.
 B. Han et al. / Construction and Building Materials 105 (2016) 189–195
Fig. 3. Loading equipment.
and adjusted by a leveling instrument. When the test set-up for
specimens was complete, loads were imposed. First the specimens
were loaded slowly to 20% of the desired load. If the ratio of the dif-
ference and the average value of the steel rods’ deformation were
greater than 10%, the specimens were unloaded and adjusted.
The specimens were reloaded, and the pressure sensor and foil
gauges were read at the same time to ensure uniform loading.
When the indication of pressure sensor reached the desired value,
the load was held and the strain of steel rods was recorded. After
that, the nuts on the middle plate were tightened. After about six
hours the stress was adjusted according to the reading of the foil
gauges on steel rods if there was a stress relaxation. At the same
time, the reference unloaded specimens were removed from the
water bath so that they had the same cured condition as the loaded
specimens.
The measuring method for compressive strength is based on the
Chinese standard [21]. The equipment used to execute the
mechanical test is electrohydraulic servo controlled pressure test-
ing machine. The maximum test pressure of the equipment is
1000 kN, and precision of the equipment is ±1%. Automatic loading
was used to make sure the loading protocol was a continuous and
uniform process, and the loading velocity was set as 0.5 MPa/s.
Fig. 4. Comparison of compressive strength development of concrete under
different loading conditions.
193
3.4. Experimental results
In Fig. 4 the experimental data of strength development of the
loaded and non-loaded specimens is shown. All of specimens’
strengths were increasing with time. For the non-loaded speci-
mens, the strength for each of those at an age of 14 days, 28 days,
and 60 days increases by 18.99%, 33.65%, and 42.55% when com-
pared to that at an age of 7 days, respectively. The development
of strength before 28 days was rapid and then the strength devel-
opment continued but became slower. Comparatively, the strength
for each of the loaded specimens at an age of 14 days, 28 days, and
60 days increased by 21.39%, 37.02%, and 46.63% when compared
to that at an age of 7 days.
As shown in Fig. 4, the strength of each of the loaded specimens
is higher than that of the corresponding non-loaded specimens.
The strength of each of the loaded specimens at an age of 14 days,
28 days, and 60 days is 2.02%, 2.52%, 2.87% higher than that of the
non-loaded specimens, respectively. Obviously, concrete under
low stress level is strengthened slightly. This strengthening trend
is fast before the age of 28 days and becomes slow after that time.
It can be seen that this value would be astringent with increasing
the time, which agrees with reference [6].
3.5. Discussion
Development of concrete mechanical properties under sus-
tained loads, coupling hydration, creep and damage, is a complex
process. While a force is applied to concrete, its physical and chem-
ical characteristics in microstructure will change. Differing from
effects of high stress levels (more than 70% compressive strength)
that induces unstable damage, one of the low level stress effects on
concrete would make it solid and dense [22], and mechanical prop-
erties of concrete are enhanced. The denser concrete and micro
cracks in concrete under sustained loads co-exists [5]. According
to the experiments, the solidification effect overwhelms the dam-
age effect from the point of view of low level stress, so the strength
increases.
The enhancement could be explained from the changes of phys-
ical concrete microstructure and chemical hydration when con-
crete is under stress. First, from a physical point of view, stresses
promote concrete solidified. Since aggregates are hard to compress,
the effect mainly occurs in cement mortar. Shah and Chandra [5]
found strength enhanced effect of concrete came from the
secondary adhesive strength of cement gel, which is caused by
194 B. Han et al. / Construction and Building Materials 105 (2016) 189–195

reducing distance between the aggregates due to the solidification.

Wittmann [23] also corroborated to this viewpoint, and he
believed the bond between cement particles will increase with act-
ing force growing proportionally. Second, from a point of view of
chemistry, stresses propel cement hydration process. As known,
once the stress and temperature were given, solubility of a kind
of solid in water should be limited to a certain concentration.
Un-hydrated cement accelerates to dissolve into water due to
stress action, and this generates a supplement process to hydration
[6], inducing strength increase. To prove that stresses promote
hydration, Coutinho [6] studied loaded and non-loaded concrete
by thermogravimetric analysis and X-ray diffraction, and found
that when heating the concrete to 600 °C, loaded concrete had
higher weightlessness rate, which means chemical-bonded water
content in cement matrix under sustained load was higher than
that without imposed stress, and a higher peak value of Ca(OH)2
diffracted wave was observed, i.e. higher content of free Ca(OH)2
was present in the concrete under stress [6]. All the evidences have
illustrated that hydratable content in concrete under sustained Fig. 6. Fitting p and q by using calculating values of concrete matrix strength and
experimental data of compressive strength.
load is higher than that without stress, and stresses provide an
extra ‘maturity’ to concrete.
corresponding aggregate fraction must be obtained. The parameter
4. Parameter fitting g can be calculated by the mix proportion of the concrete, but g⁄
cannot be calculated directly by CPM because the residual packing
4.1. Coefficients A, B, P density of each aggregate fraction is unknown. A simplified for-
mula [24] is used to calculate g⁄. For crushed aggregate
The strength of the cement matrix in an unstressed state will be
0:19
Dmin
fitted first. The related parameters are A, B and P. Test data of g  ¼ 1  0:45 ð10Þ
strength varying with age in an unstressed state is used. The 28- Dmax
day activity coefficient is 0.5 for fly ash [17]; therefore, only A
in which Dmin and Dmax are the minimum and maximum grain size
and B need to be determined.
(mm), respectively.
To ensure only A and B are unknown in the fitting formula, rel-
The strength of the concrete matrix can now be achieved. The
ative strength is used to offset the effects of factors p and q on the
coefficient p and q are fit in a nonlinear approach; according to
aggregate properties. Here the influence of the aggregate on the
experimental data, the fitted values are p = 1.33533, q = 0.00771
concrete strength is supposed to be linear. The regression values
(Adj. R2 = 0.99851), respectively. The fitted curve is as shown in
of A, B, and P are A = 0.12201, B = 2.75883, P = 0.5  B lg 28 (Adj.
Fig. 6.
R2 = 0.91807), respectively.
The regressive curve is as shown in Fig. 5.
4.3. Coefficients r0, c
4.2. Coefficients p, q
Lastly, the formula for compression strength development of
The formula for compression strength of concrete in an concrete under sustained load was fit with the stress related coef-
unstressed state is fitted in which p and q are involved. Prior to ficients r0 and c. On the condition that coefficients mentioned
calculating the strength of the concrete matrix, the MPT above are known, the least square method can be used to fit the

Fig. 5. Fitting curve of relative compression strength of concrete under sustained Fig. 7. Fitting curve of compressive strength development formula of concrete
load. under sustained load.
 B. Han et al. / Construction and Building Materials 105 (2016) 189–195 195

Table 3 51078027). And all of authors are appreciated to the valuable sug-
Comparison of calculated values by model and tested results of strength. gestions provided by Prof Mo Y.L., University of Houston, USA.
Stress ratio Age Calculated values Tested results Error
(%) (day) (MPa) (MPa) (%) References
0 7 41.4 41.6 0.49
0 14 49.8 49.5 0.71 [1] A. Oner, S. Akyuz, R. Yildiz, An experimental study on strength development of
0 28 55.5 55.6 0.17 concrete containing fly ash and optimum usage of fly ash in concrete, Cem.
Concr. Res. 35 (6) (2005) 1165–1171.
0 60 59.2 59.3 0.14
[2] A.M. Neville, Properties of Concrete, Pitman, London, 1973.
20 14 51.1 50.5 1.16
[3] G.W. Washa, P.G. Fluck, Effect of sustained loading on compressive strength
20 28 56.9 57.0 0.21
and modulus of elasticity of concrete, ACI J. 46 (5) (1950) 693–700.
20 60 60.6 61.0 0.59 [4] R.E. Davis, H.E. Davis, Flow of concrete under the action of sustained loads, ACI
J. 27 (3) (1931) 837–901.
[5] S.P. Shah, S. Chandra, Fracture of concrete subjected to cyclic and sustained
loading, ACI J. 67 (1) (1970) 816–827.
coefficients in Eq. (9) directly on the basis of experimental data, [6] A.S. Coutinho, A contribution to the mechanism of concrete creep, Mater.
and the fitted values of r0 and c at the same time, the reference Struct. 10 (1) (1977) 3–16.
unloaded specimens were removed from the water bath so that [7] D.J. Zhang, Y.F. Wang, Y.S. Ma, Compressive behaviour of FRP-confined square
concrete columns after creep, Eng. Struct. 32 (8) (2010) 1957–1963.
they had the same cured condition as the loaded specimens. The [8] K. Van Breugel, Simulation of hydration and formation of structure in
fitted values are r0 = 50.28, c = 0.00155 (Adj. R2 = 0.99141), respec- hardening cement-based materials, Delft University of Technology, Delft,
tively. The fitted curve is shown in Fig. 7. 1991. thesis for doctoral degree.
[9] K. Van Breugel, Development of temperature and properties of concrete as a
By using the fitted values of the coefficients, calculating results function of the degree of hydration, in: Proceedings: RILEM International
obtained by Eq. (9) and the corresponding experimental data are Conference on Concrete at Early Ages, Paris, French, RILEM Publication, 1982,
listed in Table 3. Comparison of these data illustrates that the pp. 179–185.
[10] D. Mikulic, I. Gabrijel, B. Milovanovic, Prediction of concrete compressive
results from Eq. (9) agree with the experimental data well, and strength, in: Proceedings: International RELEM Symposium on Concrete
almost all of error is less than 1%. It also indicates that the Modeling, Delft, Netherlands, RILEM Publication, 2008, pp. 503–510.
above-mentioned hypotheses are reasonable and suitable to the [11] A.G.A. Saul, Principles underlying the steam curing of concrete at atmospheric
pressure, Mag. Concr. Res. 2 (6) (1951) 127–140.
practice.
[12] J.M. Plowman, Maturity and strength of concrete, Mag. Concr. Res. 8 (22)
(1956) 13–22.
5. Conclusions [13] C.F. Kee, Relation between strength and maturity of concrete, Proc. ACI J. 68 (3)
(1971) 196–203.
[14] N.J. Carino, Maturity functions for concrete, in: Proceedings: RELEM
The work presented in this paper focused on establishing a new International Conference on Concrete at Early Ages, Paris, French, RILEM
model for the development of strength of concrete under low-level Publication, 1982, pp. 123–128.
[15] R.C. Tank, N.J. Carino, Rate constant functions for strength development of
sustained load based on CPM, which can take account of influences concrete, ACI Mater. J. 88 (1) (1991) 74–83.
of stress, loading age and fly-ash content. Coefficients in the mod- [16] S. Popovics, History of mathematical model for strength development of
els are fit in accordance with the experimental data. The results Portland cement concrete, ACI Mater. J. 95 (5) (1998) 593–600.
[17] F. De Larrard, Concrete Mixture Proportioning: A Scientific Approach, E & FN
obtained by the proposed models have a good match with the
Spon, London, 1999.
experimental data. [18] T.C. Powers, T.L. Brownyard, Studies of the physical properties of hardened
Considering that concrete in structures works accompanying Portland cement paste, ACI J. 18 (1946) 101–132.
[19] F. De Larrard, A model the compressive strength of structural fly ash concrete,
with action of loads, the authors suggest more researches should
ACI SP. J. 153 (6) (1995) 99–108.
be done on concrete under load effects, rather than only the con- [20] General Administration of Quality Supervision, Inspection and Quarantine of P.
crete material itself. Meanwhile, modern concrete development R. China/Standardization Administration of P.R. China, Fly ash used for cement
presents multiformity of components, for example, incorporating and concrete (GB/T 1596-2005), China Standard Press, Beijing, 2005.
[21] Construction Ministry of the P.R. China/General Administration of Quality
discrete fibers in the cement matrix. Compared with ordinary con- Supervision, Inspection and Quarantine of P.R. China, Standard for test
crete, fiber reinforced concrete was composed by three phases of methods of mechanical properties on ordinary concrete (GB/T 50081-2002),
material, fiber, cement matrix, and aggregate. Lanzoni et al. [25] China Architecture and Building Press, Beijing, 2003.
[22] B.P. Hughes, J.E. Ash, Some factors influencing the long term strength of
stated that the presence of discrete fibers can significantly improve concrete, Mater. Struct. 3 (2) (1970) 81–84.
the performance of concrete. The applicability of the proposed [23] F.H. Wittmann, Experiments to determine Van De Waals forces, in: M. Te’eni
model on FRC should also be studied while the characters in the (Ed.), Structure, Solid Mechanics and Engineering Design in Civil Engineering
Materials, Wiley, London, 1971, pp. 295–300.
hydration process of FRC should be investigated. [24] F. De Larrard, A. Belloc, The influence of aggregate on the compressive strength
of normal and high-strength concrete, ACI Mater. J. 94 (5) (1997) 417–426.
Acknowledgements [25] L. Lanzoni, A. Nobili, A.M. Tarantino, Performance evaluation of a
polypropylene-based draw-wired fibre for concrete structures, Constr. Build.
Mater. 28 (1) (2012) 798–806.
The authors would like to gratefully acknowledge the financial
support of National Science Foundation of China (Grant No.