Construction and Building Materials 163 (2018) 949–959

Contents lists available at ScienceDirect

Construction and Building Materials

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

Effects of pore structure and water absorption on internal curing

efficiency of porous aggregates
Dinghua Zou, Kun Li, Weidong Li, Haiyan Li ⇑, Tianlu Cao
School of Materials Science and Engineering, Henan Polytechnic University, Jiaozuo, Henan 454000, PR China

h i g h l i g h t s

Water release rate of aggregate is almost constant at invariant RH.

Pore structure has intensity effect on equilibrium water in aggregate, but no impact on water release rate.

Water release rate has linear relationship with reciprocal of volume water absorption.

Water release rate reflects effect of surface area on efficiency of internal curing aggregate.

In desorption test, denser container and piling of particles should be avoided.

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

Article history: This paper presents the results of experimental investigation on desorption behavior and internal curing
Received 8 June 2017 efficiency of porous aggregates with same size fraction from 2.36 to 4.75 mm, but different pore structure
Received in revised form 25 October 2017 and water absorption. Water release rate and equilibrium remaining water of aggregates at different rel-
Accepted 25 December 2017
ative humidity were obtained in desorption test. Internal relative humidity, autogenous deformation, and
compressive and flexural strength of plain and internally cured mortar were measured. Based on the
analysis of results, it was revealed that, pore structure of the aggregate has intensity effect on equilibrium
Keywords:
residual water, but indistinct on water release rate. The rate is linearly proportional to the reciprocal of
Internal curing
Aggregate
volume water absorption of the aggregate. Pore structure and volume water absorption influence the
Water desorption internal curing efficiency of aggregates. Bigger pores and lower volume water absorption of aggregate
Cementitious materials facilitate the internal curing. Therefore, water release rate, as well as equilibrium remaining water, could
Pore structure be used to characterize the internal curing efficiency of aggregates.
Relative humidity Ó 2017 Elsevier Ltd. All rights reserved.
Autogenous deformation

1. Introduction Improvement of compressive strengths of mortar accompanying

As the processing of the researches on internal curing, many
kinds of porous aggregates were taking as the water reservoirs to
provide the additional water for mitigating the self-desiccation
and high shrinkage in low w/c concrete and mortar [1]. Expanded
shale [2–7] and clay [8–12] are the most common ones. Slag [13],
bottom ash [14], and pumice [15–17] were also used. However, the
decrease of the compressive strength of internally cured concrete
and mortar at early age were usually observed when they were
employed [8,18–20]. Porous aggregates with higher densities than
lightweight aggregate, for examples the waste ceramic aggregate
[21] and normal weight porous aggregate (NWPA) [22], were
developed to eliminate the descent of compressive strength.
⇑ Corresponding author.
E-mail addresses: zoudinghua@hpu.edu.cn (D. Zou), lihaiyan@hpu.edu.cn (H. Li).
them was observed.
Based on the idea that cement reaches the possible maximum
degree of hydration, equation was derived to calculate the content
of internal curing agent and proved to be effective [23]. What
makes it complicate is that, the disparity of the properties of these
aggregates results in the different desorption behavior and thus
the volume deformation of internally cured concrete and mortar.
The compare between natural pumice and artificial expanded clay
and slate revealed that [8], former one has higher water absorption
and faster water release speed than latter aggregates due to its
higher open porosity and interconnectivity of pores. NWPA
showed the lower internal curing efficiency than expanded clay
but faster water absorbing [17]. Slag aggregates [13] obtained from
different producing processes also showed the distinction on des-
orption behavior. The shrinkage of mortar accompanying low effi-
ciency slag is contrary to the expansion of the counterpart. The

https://doi.org/10.1016/j.conbuildmat.2017.12.170
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
950 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959

compressive strength is also lower due to the insufficient internal
curing. Residual water was used as the index to evaluate the effi-
ciency of internal curing agent. In ASTM C 1761/C1761M [24], it
was suggested that the internal curing agent should lose more than
85% of the water it absorbed at the relative humidity (RH) of 94%.
To clarify the effect of properties of porous aggregate on its
internal curing behavior, water desorption of 5 porous aggregates
were surveyed. Mercury intrusion porosimetry (MIP) and scanning
electron microscope (SEM) were used to reveal the diversity of
pore structure of the aggregates. Residual water and water release
rate were analyzed to reveal the impact of pore structure and
water absorption on desorption behavior of aggregates. Internal
relative humidity, autogenous deformation, and compressive and
flexural strength of plain mortar and internally cured mortars were
measured. This paper aims to provide a more comprehensive
understand on influence of desorption behavior of porous aggre-
gate on its internal curing efficiency.
2. Materials and methods
In mixture design, w/c and fine aggregate volume fraction of all mortars were
fixed as 0.3 and 55%, respectively. Content of internal curing aggregates were calcu-
lated according to Bentz’s method [23]. Mixtures of mortar are presented in Table 4.
Bottom ash and shale did not used in mortar due to their low water absorption.
2.3. Experiments
2.3.1. Water desorption test
A setup, just as what shown in Fig. 2, was used in desorption test. In desorption
test, eight salt solutions of K2SO4, KNO3, BaCl2 2H2O, KCl, KBr, NaCl, KI, NaBr were
used to maintain the RH in 8 boxes respectively. As described in ASTM E104 [25]
and reference [9], equilibrium RH of saturated solution of these salts are 97%,
93%, 90%, 85%, 81%, 75%, 70%, and 59%, respectively. Nylon sieves were used to con-
tain the samples. Before test, sieves were placed in the box to reach the constant
weight. Appropriate amount of sample in SSD condition was placed on mesh to pre-
vent the particles from piling. The samples at 97% and 93% RH were weighted every
4 h. And the rest were weighted every 1 h. An electric balance, accurate to 0.01 g,
was used in test. Room temperature was 23 ± 1 °C.
2.3.2. Characterization of aggregate pore structure
Mercury intrusion porosimetry (MIP) of aggregate was carried out by an AutoP-
ore IV 9500 porosimeter. A Carl Zeiss Merlin Compact scanning electron microscope
was used to investigate the microstructure of aggregates.

2.1. Aggregates for desorption test

Fig. 1 shows the aggregates. Five artificial aggregates, including 2 clays, NWPA,
shale, and coal bottom ash were employed. Clay 1 was produced in Zhenjiang,
Jiangsu Province, China. Clay 2 was obtained from the particles of crushed clay
brick, which was fabricated in Jiaozuo, Henan Province, China. Shale was manufac-
tured in Yichang, Hubei Province, China. Coal bottom ash was acquired from Xin-
gyang power plant, which locates in Zhengzhou, Henan Province, China. To
diminish the influence of gradation, particles with single size fraction from 2.36
to 4.75 mm were employed. Saturated surface dry (SSD) density and water absorp-
tion of the aggregates are shown in Table 1.

2.2. Materials and mixtures of mortars

A class 42.5 Portland cement, according with Chinese national standard GB 175-
2007, was taken as binder. Chemical compositions and properties of cement are
presented in Tables 2 and 3. Natural river sand was employed as normal fine aggre-
gate. SSD density and 24 h water absorption of sand are 2.57 g/cm3 and 2.06%, Fig. 1. Image of aggregates.
respectively. Polycarboxylate Superplasticizer was employed as water reducer.

Table 1
SSD density and water absorption of aggregates.

No. Clay 1 Clay 2 NWPA Shale Bottom ash

3
SSD density/g/cm 1.27 2.12 2.23 1.52 1.84
Mass water absorption at 24 h/% 30.8 14.7 23.6 4.8 4.7
Volume water absorption at 24 h/g/cm3 0.299 0.272 0.425 0.07 0.083

Table 2
Chemical compositions of cement mass%.

CaO SiO2 Al2O3 SO3 MgO Fe2O3 K2O TiO2 Loss in ignition (950 °C)
49.73 24.14 9.53 5.67 3.40 2.68 0.86 0.28 3.4

Table 3
Properties of cement.

Specific surface area (m2/kg) Initial setting time (min) Final setting time (min) Compressive Flexural strength
strength (MPa) (MPa)
3d 28 d 3d 28 d
391 143 203 30.2 49.5 6.2 9.4

Table 4
Mixtures of mortar kg/m3.

No. Cement Water Sand (SSD) Clay 1 (SSD) Clay 2 (SSD) NWPA (SSD) Water reducer
Plain 728 218 1413 – – – 3.5
Clay 1 728 218 1078 165 – – 3.5
Clay 2 728 218 1045 – 303 – 3.5
NWPA 728 218 1177 – – 203 3.5
 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959
Fig. 2. Setup for desorption test.
2.3.3. Mortar properties test procedure
Samples with dimension of 4 cm  4 cm  4 cm, 4 cm  4 cm  16 cm were
prepared for compressive and flexural strength test, respectively. Test committed
according to Chinese national standard GB/T 17671-1999. Load rates for compres-
sive and flexural strength were 0.5 and 0.1 MPa/s, respectively.
To obtain accurate internal RH of mortar, Rotronic HygroClip HC2A-S, with the
precision of 0.8%, was adopted. Fresh mortar was cast in 10 cm  10 cm  10 cm
cubic mold. Plastic sheet was used to seal the mortar. A plastic tube was inserted
into the sample in casting. Sensor was inserted in the tube to measure internal
RH of mortar at age of 10 h. Aluminum foil was used to seal the gap between sensor
and tube.
In autogenous deformation test, corrugated tubes with a length of 400 mm and
a diameter of 30 mm were used. Deformation of samples were measured by elec-
tronic micrometer with a precision of 0.001 mm and recorded by a computer every
30 s.
3. Results and discussions
3.1. Pore structure of aggregates
SEM images in Fig. 3 show the different morphology of the
pores in aggregates. Smooth inner walls of the clay 1, bottom ash
and shale are consequent upon the melting of them at high tem-
perature. However, in production of NWPA, the raw materials did
not melt in the sintering. Coarse wall was formed. Part of the clay
2 melted in sintering, which results in a wall of pore in between.
Clay 1, bottom ash and shale experienced the similar process of
production. Nevertheless, pores in them are fairly different. There
are many big and closed pores accompanying much thin bubbles
in bottom ash and shale. Although being porous, open and inter-
connected pores in them are rarely seen. In clay 1, the ratio of
closed pores is lower than that in bottom ash and shale. This is
the reason of the low water absorptions of bottom ash and shale
and a high one of clay 1. Due to the various manufacture proce-
dure, close pores could scarcely be seen in clay 2 and NWPA. They
have the high water absorption and SSD density at the same time.
Fig. 4 displays the MIP test results of clay 2, NWPA and shale.
The results of clay 1 and bottom ash does not give here due to that
they collapsed in the test. Obvious distinction among the pore
structures of the aggregates was seen. The size of most pores in
NWPA is from 100 nm to 2 mm. The range of dimension distribu-
tion of pores in clay 2 is wider than that of NWPA. Considerable
amount of pores in clay 2 is smaller than 100 nm. The size of most
open and interconnected pores in shale is less than 100 nm. The
pores in shale are anticipated to be the smallest in three
aggregates.
Due to the absence of MIP test results, a direct compare among
clay 1, bottom ash and the rest aggregates is not practicable. How-
ever, the SEM images of the aggregates could provide the qualita-
tive information about the pore size distribution of clay 1 and
bottom ash. To promote the reliability of comparison, more SEM
images of clay 1 and bottom ash are given in Fig. 5. The similar pore
morphology and volume water absorption of shale and bottom ash
suggest the analogous pore structure of them. In SEM image, there
951
are many visible pores in clay 1. Most pores are bigger than 10 mm.
Nevertheless, there are only several visible pores in SEM image of
clay 2. Since the volume water absorption of clay 2 is only slightly
lower than that of clay 1, it could be concluded that most pores in
clay 2 are far smaller than those in clay 1. This is consistent with
the MIP test results.
3.2. Desorption behavior of aggregates
3.2.1. Water desorption of aggregates
The desorption test results of aggregates at different RH are
shown in Fig. 6. It could be seen that there are two obvious stages
of the water release of all aggregates. In first stage, water in aggre-
gates decreases linearly with the age. This was also found in des-
orption test of SAP [26]. When residual water approaches its
equilibrium amount, then a moderate stage begins. The ongoing
of second stage lasts for a much shorter period than stage one
and then desorption comes to its end.
The water release rates (slop of the linear section of desorption
curve) are affected by RH and the kind of the aggregates. The
curves become more and more steep with the reducing of RH, thus
the water release rates increase. Water release rates of shale and
bottom ash are quite higher than that of clay 1, clay 2 and NWPA.
The process of desorption reveals difference among the aggregates.
3.2.2. Equilibrium water
Fig. 7 shows the equilibrium water in aggregates at different
RH. Residual water decreases when RH descends. Besides, it varies
with the type of aggregates. Equilibrium water in clay 1 is the low-
est one, and then the NWPA and bottom ash. The residual water in
clay 2 is higher than that in shale at the 97% and 93% RH. It is con-
trary at the rest RH.
Equilibrium residual water is related to the pore structure of
aggregate. More small pores imply the more residual water. MIP
test indicates that, the percent of big pores in NWPA is higher than
that in clay 2 and shale. On the other hand, shale has the highest
ratio of thin pores in the aggregates. This results the more equilib-
rium water in shale and less in NWPA.
Another interesting phenomenon is that, compared to that in
reference [22], much lower remaining water in NWPA is seen in
this research. It could attribute to the sieve used in test. Water des-
orption would be hindered by the pile of the aggregate particles
and the denser glass dish. That should be avoided in test. In this
research, the single layer particles and the sieve facilitate the evap-
oration of water.
3.2.3. Water release rate
Water release rates of the aggregates at different RH are shown
in Fig. 8. It indicates that, RH plays a very important role in water
desorption. Water release rate increases as the RH decreases. The
trying of fitting suggests the multinomial correlations between
the water release rate and the RH, rather than the linear relation-
952 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959
(a) clay 1
(c) NWPA
(e) shale
Fig. 3. SEM images of the aggregates (500).
ship in desorption of SAP [26]. The fitting results of clay1, clay2,
NWPA, shale and bottom ash are presented as the Function 1–5,
respectively.
y ¼ 8:4772x3 þ 11:096x2  59:477x þ 39:92 R2 ¼ 0:9962
y ¼ 139:91x3  306:78x2 þ 191:04x  23:651 R2 ¼ 0:9913
y ¼ 150:01x3 þ 376:84x2  324:86x þ 97:745 R2 ¼ 0:9984
y ¼ 587:57x3 þ 1515:6x2  1353:4x þ 426:28 R2 ¼ 0:9918
y ¼ 163:23x3 þ 453:8x2  468:77x þ 180:35 R2 ¼ 0:9912
The aggregates could be classified into the fast-release and
slow-release groups according to their water release rate. Shale
and bottom ash belong to the fast-release group, and the clay 1,
clay 2 and NWPA are assign to the low-release group. In fast-
(b) clay 2
(d) bottom ash
release group, the water desorption rates of shale and bottom
ash are close to each other. It is slightly higher for shale. Neverthe-
less, the difference between the water release rates of the shale
and bottom changes with the RH. When RH is more than 81%, there
ð1Þ is a tiny difference. The difference increases rapidly with the
decrease of RH when it is lower than 81%. In slow-release group,
water release rate of clay 1 is slightly lower than that of clay 2.
ð2Þ
NWPA has a rather lower rate than clay 1 and 2.
Fig. 9 indicates the linear relationship between water release
ð3Þ rate and reciprocal of volume water absorption (VWA) at all RHs.
Table 5 presents the results of fitting. It suggests that, the water
release rate is inversely proportional to VWA.
ð4Þ
If the particle of aggregate was assumed as spheres, reciprocal
of VWA could be described as the function 6.
ð5Þ
1 4pr3 =3
¼ ð6Þ
VWA mw
where r is the radium of the particle; mw is the mass of water par-
ticle absorbed. Then function 6 could be converted to 7.
 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959
Fig. 4. MIP equivalent pore size distributions of clay 2, NWPA and shale.
1 Sr
¼ ð7Þ
VWA 3mw
where S is the surface area of the particle.
Function 7 reveals that reciprocal of VWA is proportional to the
ratio of surface area to water content when radium of particle is
constant. In this research, the particles with the single size fraction
from 2.36 to 4.75 mm are used in test to diminish the impact of
gradation on desorption of aggregate. Thus the reciprocal of VWA
is equal to the ratio of surface area to water content in some sense.
High ratio of surface area to water content promotes the speed of
Fig. 5. Random SEM images of clay 1 and bottom ash (500) (a and b are images of clay 1; c and d are images of bottom ash).
953
water release. This demonstrates the influence of the surface area
of aggregates on water desorption.
Pore structure of particles has no effect on water release rate.
Clay 1 and 2 have the similar water release rate although the pore
structures of them are different. The VWA of clay 2 is slight lower
than that of NWPA. Thus the water release rate of former is a bit
higher than that of latter. This reveals that water release rate is
only decided by the VWA. It is the same for the rest.
The end time of water release are shown in Fig. 10. It could be
seen that desorption of aggregate with higher water release rate
finishes earlier. Shale and bottom ash, which belong to the fast-
release group, are rather quicker than the slow-release group
aggregates to reach the equilibrium condition. NWPA is the last
one due to the lowest water release rate.
3.3. Properties of mortar
3.3.1. Internal RH
Development of RH in mortars with age is presented in Fig. 11.
At the age from 12 to 20 h, internal RHs of mortars are stable. The
RH in plain mortar is slightly higher than those in internally cured
mortar. Then continuous decrease of RH occurs. Plain mortar expe-
riences a steep descend of RH from 20 to 40 h. The fall of RH in
internally cured mortars is relatively mild at the same age. Differ-
ences between RHs of plain and internally cured mortars rise
rapidly with the time in this stage. Then they increase gently till
the end of the test.
It also could be seen that RHs in three internally cured mor-
tars are different. The RH in Clay 1 is obviously higher than
those in Clay 2 and NWPA. This is benefit from the big pores
in clay 1. Volume water absorption also shows its effect on
internal curing. RH in Clay 2 is higher than that in NWPA
although the internal curing aggregates in former has more fine
pores. It attributes to the higher surface area to water content
ratio of clay 2.
954 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959

3.3.2. Autogenous deformation
Autogenous deformation of mortars is displayed in Fig. 12. The
high shrinkage of plain mortar could be seen. Three internally
cured mortars show the expansion or low shrinkage in deforma-
tion test. The mortar accompanying clay 1 expands from the begin-
ning of the deformation. Mortars containing clay 2 and NWPA
shrinkage at first and then expand. After the peak, expansion of
three mortars all decrease.
Since internal curing was developed to eliminate the high auto-
genous shrinkage and prevent the early age cracking of low w/c
cementitious materials, low shrinkage or high expansion of mortar
which was designed as Bentz’s method [22] means the high inter-
nal curing efficiency of porous aggregate it contain. Autogenous
deformation of mortar in Fig. 12 implies the different efficiency
of aggregates. Water release rate of clay 2 is slight higher than that
of clay 1. However, internal curing efficiency of former is lower
than latter. This reflects the influence of pore structure on internal
curing. In reference [27], concrete with lightweight aggregate
made from fly ash has higher expansion than that with expanded
shale. It also could attribute to the more big pores in fly ash
lightweight aggregate. Compare between clay 2 and NWPA
demonstrates that the water release rate of aggregate is also
important. Although it has the thinner pores than NWPA, clay 2
has the higher efficiency. This is caused by the higher surface area
to water content ratio of clay 2.
3.3.3. Mechanical properties
Mechanical properties of mortars are presented in Fig. 13. It
could be seen that compressive strength of Clay 2 and NWPA are
higher than that of plain one at 7 and 28 days, but lower at 3 days.
It is the results of internal curing. Clay 1 has the lowest compres-
sive strength at all ages due to its high porosity. The higher com-
pressive strength of clay 2 than NWPA also should be ascribed to
the effect of internal curing. The higher internal curing efficiency
of clay 2 causes the more dense paste in Clay 2 due to the enhanced
hydration.
Flexural strength of internally cured mortar is lower than
that of plain one. It could be deduced that the strength of
aggregates has the intensity effect on flexural strength. The
impact of internal curing is also distinct. Clay 2 has a lower
flexural strength than NWPA at 3 days, whereas higher one at
7 and 28 days.

(a) clay 1

(b) clay 2
Fig. 6. Normalized remaining water in aggregates at different ages.
D. Zou et al. / Construction and Building Materials 163 (2018) 949–959 955

(c) NWPA

(d) shale

(e) bottom ash

Fig. 6 (continued)
956 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959

(a) Equilibrium water absorption (b) Normalized remaining water

Fig. 7. Equilibrium water at different RH.

Fig. 8. Water release speed of aggregates at different RH. Fig. 9. Relationship between water release rate and water absorption in volume.

Impact of internal curing on strength of low w/c mortar lies in

that, porous aggregate tends to decrease the strength, and mean- Table 5
while enhanced hydration induced by internal curing is beneficial Fitting results between water release rate and reciprocal of VWA.
to strength. For porous aggregate, low porosity always means RH Coefficient Interception R2
higher strength of it. A mortar accompanying denser porous aggre-
97 0.395 0.784 0.996
gate would have higher strength. Influence of enhanced hydration 93 0.480 0.447 0.997
is depends on age and efficiency of porous aggregates. Later age 90 0.470 0.171 0.987
and higher efficiency of porous aggregate suggest there is more 85 0.711 0.313 0.998
internal curing water in paste which could reacts with anhydrous 81 0.939 0.266 0.961
75 1.02 1.49 0.957
cement particles. Strength of mortar increases consequently. Com-
70 1.35 1.78 0.979
petition among the 3 factors influences the strength of mortar. 59 2.23 2.22 0.996

3.4. Discussions
summarized the previous works and thought the pore structure
Bentz et al. [22] believed that the surface area of the aggregates is a more key parameter responding for the internal curing effi-
is important in internal curing. However, Zhutovsky et al. [28] ciency of porous aggregates. In this research, pore structure and
 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959
Fig. 10. End time of water release.
surface area were all proved to be effective on water transportation
from internal curing aggregate to paste.
Equilibrium remaining water is related to the pore structure of
aggregate. Less remaining water implies the bigger pore in aggre-
gate and higher internal curing efficiency. It is used to reflect the
impact of pore structure on internal curing efficiency of aggregates.
However, parameter which could characterize influence of surface
area on efficiency of aggregate is absent. In this research, water
release rate was proved to be competent.
For internal curing, mass water absorption could not represent
the feature of water absorbing of aggregate. VWA is more appropri-
ate than mass water absorption to describe this.
Three potential driving forces could result in the migration of
internal curing water from aggregate to paste. Consumption of
mixing water leads to descend of RH and unsaturation of pores
in paste. Therefore, evaporation of internal curing water induced
by RH difference between aggregates and paste arises. Moreover,
due to the capillary effect, a pressure gradient with the direction
Fig. 11. Internal RH of mortars.
957
Fig. 12. Deformation of plain and internally cured mortars.
from internal curing particle to paste is established simultane-
ously. It drives the transportation of internal curing water. At last,
the interdiffusion between pore solution in paste pores and water
in internal curing aggregates also contribute to the migration due
to the concentration difference between them. However, which
one controls this process is still uncovered.
Water release driving by RH difference does not affected by
pore structure of aggregates, whereas internal curing efficiency of
aggregates is decided by pore structure when water release rate
of 2 aggregates are almost the same. It demonstrates that the cap-
illary pressure difference between the pores of internal curing
aggregate and paste dominates the water migration.
When taking only efficiency into consideration, the aggregate
with big pores is beneficial to internal curing. However, the big
pores in aggregate mean the lower strength and elastic modulus
of internally cured concrete or mortar. The particles with thin
pores and small size are more appropriate than fully graded aggre-
gate to be the internal curing aggregate.
958 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959
(a)Compressive strength
Fig. 13. Mechanical properties of mortars.
4. Conclusions
This research focus on understanding relationship between des-
orption behavior and internal curing efficiency of porous aggre-
gates. The effects of pore structure and water absorption on
water release rate, equilibrium water of aggregate, and internal
RH and autogenous deformation of internally cured mortar were
examined. The following conclusions could be drawn from the
results of the experiments and analysis.
1. Water in aggregate particles reduces linearly with the time
from the outset of desorption. A moderate stage stars when
the residual water is close to the equilibrium content. Water
release rate of internal curing aggregates is influenced by RH.
It decreases with the increase of RH. At constant RH, water
release rates of aggregates with the same size fraction are pro-
portional to their VWA. Pore structure decides the equilibrium
water in particles. However, it has few effects on water release
rate.
2. The efficiency of internally cured aggregate is depends on its
pore structure and surface area to water content ratio. Equilib-
rium remaining water represents the effect of pore structure on
efficiency of aggregate. Meanwhile, influence of surface area to
water content ratio could be reflected by water release rate.
3. Water migration from internal curing aggregate to paste is
dominated by capillary effect. Bigger pores of internal curing
aggregate induce lower negative capillary pressure in them.
Consequently, a higher capillary pressure difference between
paste and aggregate would be established. This facilitates
migration of internal curing water to paste.
4. VWA is more appropriate than mass water absorption to char-
acterize internal curing aggregate.
5. In desorption test, denser container and aggregation of the par-
ticles should be avoided.
Acknowledgement
The research in this paper is supported by National Natural
Science Foundation of China (No. U1604117 and 51678220).
(b) Flexural strength
References
[1] RILEM. Internal Curing of Concrete. RILEM Report 41, Eds. K. Kovler and O.M.
Jensen, RILEM Publications S.A.R.L., 2007.
[2] D. Cusson, T. Hoogeveen, Internal curing of high-performance concrete with
pre-soaked fine lightweight aggregate for prevention of autogenous shrinkage
cracking, Cem. Concr. Res. 38 (6) (2008) 757–765.
[3] D.P. Bentz, Influence of internal curing using lightweight aggregates on
interfacial transition zone percolation and chloride ingress in mortars, Cem.
Concr. Compos. 31 (5) (2009) 285–289.
[4] R. Henkensiefken, D. Bentz, T. Nantung, J. Weiss, Volume change and cracking
in internally cured mixtures made with saturated lightweight aggregate under
sealed and unsealed conditions, Cem. Concr. Compos. 31 (7) (2009) 427–437.
[5] R. Henkensiefkena, P. Briatka, D. Bentc, T. Nantung, J. Weiss, Plastic shrinkage
cracking in internally cured mixtures made with pre-wetted lightweight
aggregate, Concr. Int. 32 (2) (2010) 49–64.
[6] J. Browning, D. Darwin, D. Reynolds, B. Pendergrass, Lightweight aggregate as
internal curing agent to limit concrete shrinkage, ACI Mater. J. 108 (6) (2011)
638–644.
[7] R. Henkensiefken, T. Nantung, J. Weiss, Saturated lightweight aggregate for
internal curing in low w/c mixtures: monitoring water movement using X-ray
absorption, Strain 47 (2011) 432–441.
[8] B.S. Arnon, K. Konstantin, Prevention of autogenous shrinkage in high-strength
concrete by internal curing using wet lightweight aggregates, Cem. Concr. Res.
31 (11) (2001) 1587–1591.
[9] M.L. Álvaro Paul, Assessing lightweight aggregate efficiency for maximizing
internal curing performance, ACI Mater. J. 108 (4) (2011) 385–393.
[10] X. Liu, M.H. Zhang, Permeability of high-performance concrete incorporating
presoaked lightweight aggregates for internal curing, Mag. Concr. Res. 62 (2)
(2010) 79–89.
[11] F. Wang, Y. Zhou, B. Peng, Z. Liu, S. Hu, Autogenous Shrinkage of Concrete with
Super-Absorbent Polymer, ACI Mater. J. 106(2):123–7.
[12] F. Wang, J. Yang, S. Hu, X. Li, H. Cheng, Influence of superabsorbent polymers
on the surrounding cement paste, Cem. Concr. Res. 81 (2016) 112–121.
[13] M. House, C. Di Bella, H. Sun, G. Zima, L. Barcelo, W. Weiss, Influence of slag
aggregate production on its potential for use in internal curing, Transp. Res.
Rec.: J. Transp. Res. Board 2441 (2014) 105–111.
[14] M. Wyrzykowski, S. Ghourchian, S. Sinthupinyo, N. Chitvoranund, T. Chintana,
P. Lura, Internal curing of high performance mortars with bottom ash, Cem.
Concr. Compos. 71 (2016) 1–9.
[15] S. Zhutovsky, K. Kovler, A. Bentur, Influence of cement paste matrix properties
on the autogenous curing of high-performance concrete, Cem. Concr. Compos.
26 (5) (2004) 499–507.
[16] M. Sßahmaran, M. Lachemi, K.M.A. Hossain, V.C. Li, Internal curing of
engineered cementitious composites for prevention of early age autogenous
shrinkage cracking, Cem. Concr. Res. 39 (10) (2009) 893–901.
[17] B. Akcay, M.A. Tasdemir, Effects of distribution of lightweight aggregates on
internal curing of concrete, Cem. Concr. Compos. 32 (8) (2010) 611–616.
[18] K. Raoufi, J. Schlitter, D. Bentz, J. Weiss, Parametric assessment of stress
development and cracking in internally cured restrained mortars experiencing
autogenous deformations and thermal loading, Adv. Civ. Eng. 2011 (2011) 1–
16.
 D. Zou et al. / Construction and Building Materials 163 (2018) 949–959
[19] D. Zou, J. Weiss, Early age cracking behavior of internally cured mortar
restrained by dual rings with different thickness, Constr. Build. Mater. 66
(2014) 146–153.
[20] S. Zhutovsky, K. Kovler, A. Bentur, Autogenous curing of high-strength
concrete using pre-soaked pumice and perlite sand. self-desiccation and its
importance in concrete technology, in: B. Persson, G. Fagerlund (Eds.),
Proceeding of the 3rd International Research Seminar 2002 Lund 161 174
[21] M. Suzuki, M.S. Meddah, R. Sato, Use of porous ceramic waste aggregates for
internal curing of high-performance concrete, Cem. Concr. Res. 39 (5) (2009)
373–381.
[22] D. Zou, H. Zhang, Y. Wang, J. Zhu, X. Guan, Internal curing of mortar with low
water to cementitious materials ratio using a normal weight porous aggregate,
Constr. Build. Mater. 96 (2015) 209–216.
[23] D.P. Bentz, K.A. Snyder, Protected paste volume in concrete-extension to
internal curing using saturated lightweight fine aggregate, Cem. Concr. Res. 29
(11) (1999) 1863–1867.
959
[24] ASTM C 1761, C1761M-12, Standard Specification for Lightweight Aggregate
for Internal Curing of Concrete, American Society of Testing and Materials,
2012.
[25] ASTM E104, Practice for Maintaining Constant Relative Humidity by Means of
Aqueous Solutions, American Society of Testing and Materials, 2012.
[26] F. Wang, J. Yang, H. Cheng, J. Wu, X. Liang, Study on mechanism of desorption
behavior of saturated superabsorbent polymers in concrete, ACI Mater. J. 112
(3) (2015).
[27] W. Ya, X. Yaping, Z. Qianqian, Internal curing efficiency of prewetted LWFAs on
concrete humidity and autogenous shrinkage development, ASCE J. Mater. Civ.
Eng. 26 (2014) 947–954.
[28] S. Zhutovsky, K. Kovler, A. Bentur, Revisiting the protected paste volume
concept for internal curing of high-strength concretes, Cem. Concr. Res. 41 (9)
(2011) 981–986.