Materials and Design 92 (2016) 499–511

Contents lists available at ScienceDirect

Materials and Design

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

Self-healing in cementitious materials: Materials, methods and

service conditions
Haoliang Huang a,b,⁎, Guang Ye b,c, Chunxiang Qian a, Erik Schlangen b
a
School of Materials Science and Engineering, Southeast University, Nanjing, China
b
Department of Civil Engineering and Geoscience, Delft University of Technology, The Netherlands
c
Department of Structural Engineering, Ghent University, Belgium

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

Article history: In this paper, mechanisms of self-healing in cementitious materials, i.e. autogenous self-healing, self-healing
Received 15 August 2015 based on mineral admixtures, self-healing based on bacteria and self-healing based on adhesive agents, are
Received in revised form 14 December 2015 reviewed. Literature shows that all mechanisms of self-healing are effective, to some extent, under some partic-
Accepted 15 December 2015
ular conditions. It reveals that not any particular method of self-healing is the best, but one can be the most suit-
Available online 17 December 2015
able for a particular situation. For better application of self-healing concept in engineering practice, favorable
Keywords:
situations for self-healing in cementitious materials are summarized. The required environmental conditions
Self-healing for each self-healing mechanism are analyzed. Additional costs for realizing self-healing in concrete structures
Cementitious materials are also discussed. Based on the aforementioned aspects of self-healing in cementitious materials, perspectives
Mechanism for further research on application of self-healing in engineering practice are proposed.
Service conditions © 2015 Elsevier Ltd. All rights reserved.
Cost

1. Introduction
Concrete is a brittle composite cementitious material that easily frac-
tures under tensile loading. For this reason, reinforcement is installed in
order to carry the tensile cross sectional forces after cracking. From this
point of view, reinforced concrete is always designed to allow the occur-
rence of cracks. Cracks as such are not regarded as failure of reinforced
concrete as long as a prevailing crack width criterion is not exceeded
[1]. However, they provide preferential accesses for aggressive agents,
such as chlorides, sulfates and carbonates. These aggressive agents can
not only induce corrosion of reinforcement steel, but also degrade the
concrete. Thus service life of reinforced concrete structures is shortened.
In addition, cracks cause leakage in concrete structures, such as water
reservoirs, roofs and water pipes, and negatively affects their
functionality.
For solving the aforementioned service life and leakage problems
caused by cracks, man-made repair is a common solution [2]. Although
man-made repair can prolong service life of reinforced concrete struc-
tures, it has some limitations. For instance, the cost of man-made repairs
are usually very high [3]. Moreover, if structures, like bridges and tun-
nels, have to be taken out of service for repair, the indirect costs are gen-
erally several times higher than the direct costs [2]. Apart from the high
cost of repair, it is recognized that realizing durable repairs is difficult,
even though the quality of such repairs has substantially increased in
⁎ Corresponding author at: School of Materials Science and Engineering, Southeast
University, Dongnandaxue Road 2, Nanjing, China.
recent years. Most of these repairs can only last for ten to fifteen years
[2]. Furthermore, it is difficult to repair cracks which are not accessible,
such as the cracks in underground concrete structures [4].
In comparison, self-healing of cracks in concrete structures could be
beneficial [4]. Self-healing of crack blocks the crack pathway and, there-
fore, prevents from water leakage. Functionality of the concrete struc-
tures is regained. At the same time, block of crack pathway due to
self-healing inhibits ingress of aggressive ions. As a result, service life
of concrete structures is prolonged. Moreover, self-healing of crack
increases abrasion between crack surfaces and, therefore, restores
compressive strength of the cracked concrete matrix [5]. Similarly,
self-healing of damages at interfaces between fibers and cementi-
tious matrix can restore mechanical properties of fiber reinforced
concrete. Therefore, it is significant to make reinforced concrete
structures smart enough to detect their own damage and repair
themselves [6].
As a novel idea, self-healing of cracks has attracted much attention
worldwide in recent years and some reviews on self-healing have
been published, mainly focusing on healing agents and methods to
evaluate the efficiency of self-healing [6–10]. In this paper, different
mechanisms of self-healing in cementitious materials and the corre-
sponding healing agents were reviewed. For better application of
self-healing concept in engineering practice, favorable situations for
self-healing in cementitious materials are summarized. The required
environmental conditions for each self-healing mechanism are ana-
lyzed. The costs due to the addition of self-healing agents are also
discussed. Based on the aforementioned aspects of self-healing in

http://dx.doi.org/10.1016/j.matdes.2015.12.091
0264-1275/© 2015 Elsevier Ltd. All rights reserved.
500 H. Huang et al. / Materials and Design 92 (2016) 499–511

cementitious materials, perspectives for further research on application

of self-healing in engineering practice are proposed.

2. Mechanisms of self-healing and required conditions

Based on the literature survey, self-healing of cracks in cementitious

materials can be grouped into four categories according to mechanisms:
1) autogenous self-healing,
2) self-healing based on mineral admixtures,
3) self-healing based on bacteria,
4) self-healing based on adhesive agents.
The physico-chemical healing process, influencing factors, advan-
tages and disadvantages of these categories of self-healing are described Fig. 1. Autogenous self-healing due to precipitation of calcium carbonate in the presence of
in this section. water and dissolved CO2 [18].
The figure was reprinted from [9].

2.1. Autogenous self-healing

2.1.1. Phenomena of autogenous self-healing and its effect found that the main components of reaction products formed in cracks
Autogenous self-healing in Portland cement concrete has attracted in Portland cement paste with w/c ratio of 0.3 were portlandite and C–
much attention since it was observed many years ago. According to S–H. The portlandite accounted for almost 80% by mass of the reaction
Hearn [11], the phenomena of autogenous self-healing had already products. It indicates that not only further hydration of unhydrated ce-
been noticed in water retaining structures, culverts and pipes by Hyde ment is responsible for autogenous self-healing, but also the recrystalli-
[12] by the end of nineteenth century. In 1920s, a more systematical zation of portlandite leached from the bulk paste.
analysis of autogenous self-healing was reported by Glanville [13].
After that, autogenous self-healing of cracks in concrete bridges was 2.1.3. Mechanisms of autogenous self-healing
also investigated [14,15]. By summarizing the aforementioned information, it can be found
The effect of autogenous self-healing on water leaking through that the main mechanisms of autogenous self-healing could be:
cracks was extensively studied by Clear [16], Hearn [17] and Edvardsen
[18]. Moreover, Reinhardt et al. [19] correlated this effect with different (1) further hydration of unhydrated cement;
temperatures and crack widths. The reduction of chloride ingression (2) recrystallization of portlandite leached from the bulk paste;
through cracks due to autogenous self-healing was reported by Fidjestol (3) formation of calcite.
et al. [20], Ramm et al. [21] and Otsuki et al. [22]. Apart from the durabil-
ity aspect, the improvement of mechanical properties of concrete due to
autogenous self-healing has been explored as well. Lauer and Slate [23] All these physic-chemical processes can proceed simultaneously, but
demonstrated that the tensile strength measured perpendicular to the their rates are different. Further hydration of unhydrated cement at
crack plane increased after autogenous self-healing of cracks. In that crack surfaces can take place immediately after water penetrates into
study, the influence of age and curing conditions were also taken into cracks, while recrystallization of portlandite and formation of calcite
account. Similarly, the recovery in strength of concrete was also found proceed very slowly. Usually there are very few CO2− 3 ions existing in
by Dhir et al. [24], Granger et al. [25] and Ferrara et al. [5]. the bulk paste. In this case, calcite is hardly formed. However, CO23 −
ions in cracks can come from outside environment, although the diffu-
2.1.2. Reaction products of autogenous self-healing sion of CO2− 2−
3 ions in water is very slow [33]. A gradient of CO3 ion con-
Although the effects of autogenous self-healing have been investi- centration is formed within the cracks. It is conceivable that
gated for many years, reaction products of autogenous self-healing de- concentration of CO23 − ions nearby the crack mouth is higher than
tected by researchers are not consistent. Jacobsen and Sellevold [26] that inside the crack. As a result, calcite is formed first nearby the
found some newly formed C–S–H, portlandite and ettringite in cracks crack mouth, as observed by Sisomphon [30]. As self-healing proceeds,
in high performance concrete that had suffered from frost deterioration more and more CO23 − ions reach the locations inside the cracks. The
and was cured in water for 3 months (under exposure to atmosphere).
Schlangen and Ter Heide [27] detected newly formed C–S–H in cracks
after the cracked samples were cured in unsealed water reservoir for
56 days. They concluded that autogenous self-healing was caused by
further hydration of unhydrated cement clinker.
Edvardsen [18] found calcium carbonate (CaCO3) in cracks after au-
togenous self-healing. Investigations by Yang et al. [28] and Qian et al.
[29] also confirmed the existence of CaCO3 in cracks. According to
Edvardsen [18], when CO2 in the air dissolves in water, CO2− 3 ions dif-
fuse into cracks through the crack mouth. CaCO3 precipitates in cracks
when the concentration of Ca2+ and CO2− 3 ions reach supersaturation
level (see Fig. 1 from [18]). However, Sisomphon [30] reported that,
since the concentration of CO23 − nearby the crack mouth is higher
than that inside the cracks, CaCO3 tends to precipitate nearby the
crack mouth (see Fig. 2 from [30]). Parks et al. [31] got a similar conclu-
sion demonstrating that calcite was not detected in cracks in internal
concrete.
Huang [32] characterized reaction products of autogenous self- Fig. 2. Experimental evidence for preferential precipitation of calcium carbonate nearby
healing under sealed conditions by means of EDS, TGA and FTIR. They the crack mouth [30].
 H. Huang et al. / Materials and Design 92 (2016) 499–511
formed portlandite inside the cracks will be carbonated gradually, lead-
ing to more and more calcite formation. Based on these physic-chemical
processes it can be concluded that the reaction products formed in
cracks disperse heterogeneously and the components varies with
time. This explains why the reaction products of autogenous self-
healing observed by different researchers are inconsistent.
2.1.4. Effect of cement type and supplementary materials on autogenous
self-healing
It is conceivable that types of cement and supplementary materials
have significant influence on the properties of concrete. Concrete struc-
tures built in the 1970s with coarse cement were emphasized because
of good durability [34]. It may be attributed to the larger cement particle
size that increases the potential of autogenous self-healing of cracks oc-
curring in the later period of the service life of these concrete structures
[34]. More quantitative analysis on the influence of cement particle size
on autogenous self-healing is needed.
Van Tittelboom et al. [35] investigated the influence of fly ash and
blast furnace slag on the capacity of autogenous self-healing in cemen-
titious materials experimentally. They concluded that replacement of
cement by fly ash or slag improves autogenous self-healing. Similar con-
clusion was drawn by Termkhajornkit et al. [36] that fly ash increases
the self-healing capacity of concrete. According to Sahmaran et al.
[37], although samples with fly ash have more reactive materials and,
thus, a higher capacity of self-healing expected, more evident self-
healing product was observed in the sample with slag, rather than in
the sample with fly ash.
It should be mentioned that calcium hydroxide is necessary for fur-
ther reaction of slag and fly ash during the healing process [38]. As
shown in Fig. 3 from [35], if a very large percentage of cement was re-
placed with slag and fly ash, calcium hydroxide is not enough for further
reaction of slag and fly ash. In this case, the efficiency of autogenous self-
healing may be affected negatively.
2.1.5. Effect of ions from outside environment on self-healing
Ions from outside environment of structures can diffuse into cracks.
These ingressive ions change chemical potentials in the solution in
cracks. Formations of reaction products of self-healing are then affected
by these ingressive ions.
One obvious example is the transformation of portlandite into calcite
after the diffusion of CO2−3 ions from the outside environment into the
crack, which was discussed in Section 2.1.3. Actually carbonation of
portlandite facilitates the filling of cracks since the volume of calcite is
11% more than that of portlandite as a reactant [39]. When CO2− 3 ions
are present inside cracks, C–S–H may be carbonated as well. According
to thermodynamic modeling with CHES [40], calcite and silica gel are
Fig. 3. Autogenous self-healing capacity of cementitious materials with different percent-
ages of slag or fly ash [35].
501
formed after carbonation of C–S–H. Modeling results in [39] show that
the total volume of calcite and silica gel is less than that of the reactant
C–S–H. Therefore, carbonation of C–S–H has negative effect on the effi-
ciency of autogenous self-healing [39], which is different from the car-
bonation of portlandite.
Parks et al. [31] investigated the effect of ions in bulk water on autog-
enous self-healing. They demonstrated that appreciable efficiency of au-
togenous self-healing of cracks can be achieved when there is sufficient
magnesium and silicon in the water [31]. In comparison, calcium does
not promote the efficiency of autogenous self-healing [31]. However,
Huang et al. [41] compared the efficiency of self-healing in cement
paste cured in saturated calcium hydroxide solution under sealed condi-
tion with that in water. A conclusion was drawn that calcium promotes
autogenous self-healing. Recently, Palin et al. [42] compared precipi-
tates on surfaces of cement paste cured in sea water with those on sur-
faces of cement paste cured in fresh water. They demonstrated that
although precipitates characterized are on surfaces of cement paste,
rather than surfaces of cracks, the characterization results can give in-
formation of the effect of ions on autogenous self-healing [42]. They
found that the thickness of the layer of precipitates on the surfaces of
Portland cement paste cured in sea water is larger than that on cement
paste cured in fresh water [42]. However, for the cure in sea-water the
bulk matrix just below the surface where precipitates formed was
more porous, compared to the cure in fresh water [42]. It seems that
materials in the bulk matrix leached out and reacted with the ions in
sea-water, leading to a porous microstructure of bulk matrix. It is con-
ceivable that precipitations inside cracks should be more complex
than on specimen surfaces since precipitates inside cracks are related
with transport of ions from bulk sea-water into the cracks. Further re-
search is still necessary.
2.1.6. Densification of bulk matrix adjacent to the crack surfaces besides fill-
ing of the crack
As described in Section 1, self-healing of cracks leads to extension of
service life of concrete structures. It is agreed that the extension of ser-
vice life is due to the filling of cracks with reaction products of self-
healing. In addition to filling of cracks, densification of the bulk paste ad-
jacent to the crack surfaces (illustrated in Fig. 4) in the healing process
were noticed and quantified by Huang et al. [43] by means of nuclear
magnetic resonance (NMR) tests. ‘Extra water’ from the outside envi-
ronment absorbed by the bulk matrix caused additional hydration of
unhydrated cement in the bulk matrix adjacent to the crack surfaces
[39]. The denser bulk paste adjacent to the crack surfaces can decrease
the ingress of aggressive ions into the bulk matrix and slow down deg-
radation process of the concrete. This self-healing phenomena (refers to
self-healing of pores, instead of cracks) was also observed at surfaces of
concrete with low w/c ratio [44]. Different from the filling of crack with
reaction products, which is strongly related to the crack width, the den-
sification of bulk paste adjacent to the crack surfaces is rarely dependent
on the crack width. The observation and quantification of densification
of the bulk paste adjacent to the crack surfaces provides a new insight
into autogenous self-healing.
Fig. 4. Schemes for illustrating densification of bulk paste adjacent to the crack surfaces
and the surfaces of concrete [43].
502 H. Huang et al. / Materials and Design 92 (2016) 499–511
2.2. Self-healing based on mineral admixtures
2.2.1. Mineral admixtures for self-healing
Self-healing can be attributed to reactions of mineral admixtures in
cementitious materials. These mineral admixtures are added in the con-
crete mixture during mixing. After concrete cracks, some unreacted
mineral admixtures are present at crack surfaces. When water pene-
trates into the cracks, these mineral admixtures start to react with
water in the cracks. The cracks are then expected to be filled with reac-
tion products.
By now, the mineral admixtures for self-healing can be categorized
into two groups: expansive additive and crystalline additive [30]. In
the case of expansive additive, the volume of the reaction products is
larger than that of the admixture itself and the ratio of expansion de-
pends on the composition of the admixtures [45]. Regarding the crystal-
line additive, its components can react with Ca(OH)2 to form crystalline
products [30]. As shown in Table 1, Kishi and co-workers [46,47] used a
mixture of expansive agents, i.e. calcium sulfoaluminate
(Ca4(AlO2)6SO4), free lime (CaO) and anhydrite (CaSO4). In their stud-
ies, another expansive additive called geo-material, which mainly con-
sists of silicon dioxide, sodium aluminum silicate hydroxide and
montmorillonite clay, was also studied [47]. Similar to Kishi's research
in [46–48], various types of minerals as expansive admixtures for self-
healing were also investigated by Sisomphon [30]. The effects of the
mixing ratio of the mineral compositions on the capacity of self-
healing were explored as well [30]. Apart from expansive additives, a
synthetic cementitious material called “crystalline additive”, which con-
tains reactive silica and some crystalline catalysts, was also applied for
self-healing [30,49,50]. It has been reported that these mineral admix-
tures lead to remarkable improvement of water tightness of cracks
due to self-healing [5,30,46–51] (see Fig. 5 from [47]).
Using mineral admixtures as self-healing agent have some advan-
tages. For example some minerals are able to react intensively with
water, self-healing of cracks proceeds fast. Moreover, because of their
expansive character, the expansive additive can definitely improve the
efficiency of self-healing.
2.2.2. Technical problems for using mineral admixtures for self-healing
However, there are some technical problems when mineral admix-
tures are used as self-healing agent. For instance, if the minerals are di-
rectly added into the concrete mixtures without any protection, once
they come in contact with water during the mixing of concrete, they im-
mediately start to react [45]. As a result, these added minerals are con-
sumed before cracking. Moreover, when an expansive additive is used,
expansion always occurs in the interior of the concrete matrix, which
could cause damage [52]. Therefore, when mineral admixture is used
to realize self-healing, pre-processing, like encapsulation of admixture,
is necessary. The application of encapsulation techniques on self-
healing based on mineral admixtures is being investigated [53–55].
One more vital requirement for this self-healing mechanism is that
water should be continuously available in cracks, the same as
Table 1
Mineral admixtures for self-healing reported in literature.
Categorization Materials References
Expansive Calcium sulfoaluminate (4CaO·3Al2O3·SO3) [29]
Free lime [29]
Anhydrite [29]
United expansing agent [49]
Bentonite [49]
Composites of silicon dioxide, sodium aluminum [45–47]
silicate hydroxide and montmorillonite clay
Crystalline Types of carbonates, i.e. NaHCO3, Na2CO3 [46]
Talcum powder [49]
Sodium silicate [51]
autogenous self-healing. Similar to the case of autogenous self-healing,
ions coming from environment have significant effect on physico-
chemical processes of self-healing based on mineral admixtures. How-
ever, by now there are no studies reported about the effect of ions
from environment on self-healing in cementitious materials based on
mineral admixtures. Some ingressive ions may promote self-healing,
while some may inhibit. It is suggested that the effect of ingressive
ions on the healing process should be taken into account in future
research.
2.3. Self-healing based on bacteria
The idea of bacteria-based self-healing is to utilize bacteria to pro-
mote precipitation of calcium carbonate in cracks. In 1990s, Gollapudi
et al. [56] suggested to use bacteria to induce the precipitation of calci-
um carbonate (CaCO3) to repair cracks. The precipitation of calcium car-
bonate can be caused by various metabolic pathways, such as the
hydrolysis of urea and the oxidation of organic acids [4].
Compared to the other pathways for generating carbonate, the hy-
drolysis of urea has several advantages. For instance, it can easily be con-
trolled and it has the potential to produce high amounts of carbonate
within a short time [57,58]. Catalyzed by means of urease, urea is de-
graded to carbonate and ammonium, and consequently increases the
pH value, as well as the carbonate concentration in the bacterial envi-
ronment [57]. One mole of urea is hydrolyzed intracellularly to 1 mol
of ammonia and 1 mol of carbonate (Eq. 1), which spontaneously hy-
drolyzes to 1 mol of ammonia and carbonic acid (Eq. 2) [57]. These
products subsequently reach the equilibrium in water to form bicarbon-
ate and 2 mol of ammonium and hydroxide ions (Eqs. 3 and 4) [57]:
COðNH2 Þ2 þ H2 O⇒H2 COOH þ NH3 ð1Þ
NH2 COOH þ H2 O⇒NH3 þ H2 CO3 ð2Þ
2NH3 þ 2H2 O⇔2NH4 þ þ 2OH− ð3Þ
2OH− þ H2 CO3 ⇔CO3 2− þ 2H2 O: ð4Þ
In the presence of calcium ions, calcium carbonate is precipitated,
once a certain super-saturation level is reached [57]. As shown in Fig.
6, because of the negative charge of the cell wall, calcium ions are
attracted (Eq. 5). As a result, the crystals precipitate on the bacterial
cell (Eq. 6) [4,57]. In addition, precipitation also takes place in the bulk
liquid phase [57].
Ca2þ þ Cell⇒Cell−Ca2þ ð5Þ
Cell−Ca2þ þ CO3 2− ⇒CaCO3 ð6Þ
Another metabolic pathway to produce calcium carbonate is the ox-
idation of organic acids. Compared to the hydrolysis of urea, which pro-
duces excessive ammonium, the oxidation of organic acids has less
environmental impact [59]. Moreover, during the precipitation of calci-
um carbonate through this metabolic way, CO2 is produced as well. The
produced CO2 can also react with portlandite (Ca(OH)2), which is quan-
titatively an important hydration product of Portland cement, to form
more calcium carbonate [60,61].
However, in order to use bacteria to heal cracks in concrete, some
technical problems have to be solved. The bacteria should be protected
not only against the alkaline environment in concrete, but also against
the decreasing space in the matrix when hydration of cement proceeds
[4,60]. As found by Jonkers [60], when the bacteria spores, which can be
viable for up to 50 years, were directly added into concrete, their life-
time dramatically decreased to only a few months. This is caused by
the hydration of cement grains. As the cement grains hydrate, most of
pores becomes smaller than bacterium spores with the size of 1 μm,
 H. Huang et al. / Materials and Design 92 (2016) 499–511 503

Fig. 5. Experimental evidence of self-healing based on mineral admixture [47].

which causes the cell to collapse [60]. It is demonstrated that the immo-
bilization of bacteria in porous clay aggregates before the mixing of con-
crete can prolong the lifetime of bacteria enormously. Self-healing can
be triggered later on when cracks intersect these clay particles
[60–63]. Contemporarily, researchers in Ghent University are trying to
immobilize bacteria in silica gel or polyurethane [64]. Qian et al. [65]
in Southeast University tried to use sodium alginate and agar to protect
bacteria spores. The “food” for the bacteria, urea or organic acids, should
also be embedded in the matrix in similar ways. It should be mentioned
that the essential condition for bacteria-based self-healing is that water
is present in cracks [66–68].
Obviously, ions from environment also influence the precipitation of
calcium carbonate induced by bacteria. It is conceivable that Ca2+ and
CO2−3 ions from environment can promote precipitation of calcium car-
bonate. However, the influences of other ions (besides Ca2+ and CO2− 3
ions) from environment on the activity of bacteria and the capacity of
self-healing based on bacteria are still not clear.

Fig. 6. Simplified representation of carbonate precipitation induced by the hydrolysis of urea with the help of bacteria [57].
504 H. Huang et al. / Materials and Design 92 (2016) 499–511
Fig. 7. Visualization of crack-sealing with hardened adhesive agents by means of X-ray
Computed Tomography [69].
2.4. Self-healing based on adhesive agents
The hardening of adhesive agents in cracks has the potential to seal
the cracks and connect both crack surfaces (see Fig. 7 from [69]). These
adhesive agents can be either a one-component or two-component
agent, or even a multi-component agent. The hardening processes
vary with the properties of the agent. Various adhesive agents applied
for self-healing in recent studies are discussed below.
Epoxy is one of the adhesive agents often used for self-healing. In
some investigations, a two-component epoxy was encapsulated and
pre-embedded in concrete for self-healing [70–72]. When the capsules
became intersected by cracks, both components of the epoxy were re-
leased and mixed with each other. These two mixed components then
reacted and hardened to heal the cracks. However, the reaction can be
negatively influenced because the ratio of these two components
leaking into the cracks could not be controlled and the optimum mixing
could not be achieved [4]. A one-component epoxy has been reported to
be more effective for self-healing. For instance, Thao et al. [73] embed-
ded one-component epoxy in concrete by using sealed glass tubes.
After the glass tubes were broken by cracks, the one-component
epoxy leaked into the cracks. Once this type of epoxy was exposed to
air, it started to harden and finally repaired the cracks effectively.
Nishiwaki et al. [74] also reported positive results of self-healing by
using one-component epoxy.
Methylmethacrylate (MMA) is another type of adhesive agent with
two or multi components that can be utilized for self-healing. Yang
et al. [75] used microcapsules to supply methylmethacrylate and
triethylborane (TEB) to cracks for self-healing. Dry and McMillan [76]
even used three-component MMA for self-healing in concrete. In their
study, in order to supply this multi-component MMA to cracks, a
multi-channel vascular system was built inside concrete. The liquid
components were delivered to cracks through this vascular system
from the outside (more details can be seen in Section 3.2). Compared
with the two-component epoxy, MMA has a lower viscosity and thereby
the mixing of components is better. Nevertheless, because the viscosity
of MMA is low, it can easily leak out of the cracks. This leaking may de-
crease the efficiency of self-healing. To overcome this problem, poly
methylmethacrylate (PMMA) with higher viscosity was applied to re-
place MMA as a healing agent [4].
Cyanoacrylate has been also employed as adhesive agent for self-
healing. Cyanoacrylate has a low viscosity and is able to penetrate into
the matrix. As a result it can connect the crack surfaces strongly [4,77].
Furthermore, as a one-component adhesive agent, it can harden within
a short time after exposure to air [4]. Therefore, cyanoacrylate is fre-
quently used in self-healing concrete [78–84].
Apart from the aforementioned agents, other healing agents such as
silicon [85] and tung oil [3] were used [8]. In general, the efficiency of
self-healing based on adhesive agents mainly depends on the type of ad-
hesive agents. One-component agents are easier to operate with and
their efficiency is higher than that of two or multi component agents
[4]. Usually, adhesive agents make both crack surfaces bond strongly
and, therefore, lead to good recovery of mechanical properties of con-
crete [4].
In addition, one of the important requirements for this self-healing
mechanism is the successful supply of adhesive agents to cracks.
These liquid adhesive agents are generally supplied to cracks by using
capsules or vascular systems. Detailed information about the methods
to supply liquid healing agent to cracks will be discussed in Section 3.
3. Methods to supply healing agents to cracks
Except autogenous self-healing, self-healing with other mechanisms
usually requires the supply of healing agents to the cracks. By now, the
main methods to supply healing agents to cracks deliberately are by
using capsules or vascular systems. The state-and-art of these two
methods is presented below.
3.1. By using capsules
3.1.1. Encapsulation techniques
Capsules can be used to store healing agents in the matrix. Once
cracks appear and propagate through the matrix, some capsules can
be hit and ruptured by the cracks, the healing agents are released into
the cracks, providing that the healing agents are in liquid form. This con-
cept has been used for self-healing in polymers for more than 10 years
[86].
Similar to the case in polymer materials, many researchers have
used capsules to supply liquid-based adhesive agents to cracks in ce-
mentitious materials for self-healing, as mentioned in Section 2.4.
Also, solid healing agents, such as bacteria and mineral admixtures de-
scribed in Sections 2.2 and 2.3, respectively, can be encapsulated and
mixed into concrete.
There are various encapsulation techniques available so far. They can
be grouped into physical encapsulation and chemical encapsulation
[87]. Experience of encapsulation from the food industry and pharma-
ceutical industry can be useful for encapsulating healing agents for con-
crete mixtures [88,89].
In general, the use of capsules for self-healing has the advantage that
self-healing of cracks can take place automatically [86]. However, this
method has its limitation. First, since the capsules are randomly dis-
persed in the matrix, only a small part of them can be hit and ruptured
by cracks [90–92]. In this case, only small volume of healing agents can
be released. There is a probability for a crack hitting different numbers
of capsules. The probability of a crack hitting capsules depends on the
capsule dosage and size [90]. Second, even though a crack initiating in
the bulk matrix hits a capsule, it is still possible that the crack deflects
along the interface between the capsule shell and the bulk matrix,
 H. Huang et al. / Materials and Design 92 (2016) 499–511
rather than intersect the capsule. In order to guarantee that cracks are
able to intersect the capsule, mechanical properties of the capsule
shell and its bonding with the bulk matrix should meet some particular
requirements.
In the following subsection, researches of probability of a crack hit-
ting capsules are reviewed. Requirements of mechanical properties of
capsule shell and its bonding with the matrix are summarized. The effi-
ciency of supply of liquid healing agents by means of capsules is
discussed.
3.1.2. Probability of a crack hitting capsules
Currently several analytical models have been proposed to calculate
the probability of a crack hitting capsules [91,93]. In addition to the an-
alytical models, the probability of a crack hitting capsules can be deter-
mined by performing random experiments with the help of computers
[92,94]. Probability that the number of capsules hit by a crack (X) is
more than a specific value (x) is defined as [95]:
nðX ≥xÞ
P ðX ≥xÞ ¼
n presented a design of polymeric capsules that were able to resist the
where X ≥ x is a possible event that the number of capsules hit by a crack
is more than a certain number (x) and P(X ≥ x) is the corresponding
probability of occurrence of the event X ≥ x. n is the total number of ran-
dom experiments performed with a computational algorithm. n(X ≥ x) is
the number of random experiments in which the number of capsules hit
by the crack is more than a certain number x.
According to Huang et al. [92], for cement paste with capsules of a di-
ameter of 100 μm and a dosage of 1% by volume, more than 41 capsules
can be hit by a planar crack (5 mm in length and 5 mm in depth) with a
probability of 99% (P(X ≥ 41) = 99%, see Fig. 8). The probability P(X ≥ x)
decreases sharply with the increase of the number of capsules hit.
With the same probability, the number of capsules hit by a crack in-
creases as the capsule dosage increases. Results in the same study [92]
show that when the capsule dosage is the same, the total volume of cap-
sules hit by the crack increases linearly with capsule diameter, although
the number of capsules hit by a crack decreases.
3.1.3. Mechanical properties of capsules
As mentioned before, even though a crack initiating in the bulk ma-
trix hits a capsule, it is still possible that the crack deflects along the in-
terface between the capsule shell and the bulk matrix, rather than
Fig. 8. Probability that the number of capsules hit by crack is more than a certain number,x,
influenced by capsule dosage. The capsule diameter is 100 μm. The crack is 5 mm in length
and 5 mm in depth [92].
505
intersect the capsule. Whether a crack initiating in the bulk matrix can
intersect a capsule after hitting it depends on the mechanical properties
of capsules and the bulk matrix, and the bonding strength between
them [96]. It has been reported that the stiffer the capsule, the higher
bonding strength required to prevent interfacial cracks (cracks propa-
gate along interfaces between capsules and the bulk matrix) [96]. Actu-
ally, the bonding between capsule and the bulk matrix is related to
interfacial transition zone (ITZ). The ITZ between capsule and bulk ma-
trix varies with materials of capsule shell and mixture of the bulk paste.
It can be investigated with similar methods for studying the ITZ be-
tween aggregate and bulk paste.
Another important aspect about the mechanical properties of cap-
sules is the resistance of capsules against mixing forces during mixing
of concrete. Capsules made of glass, which were used mostly for inves-
tigating feasibility of self-healing based on liquid adhesive agents in the
last decade [69,80,97,98], have poor resistance against the mixing forces
during concrete mixing. Van Tittelboom [4] proposed to coat glass cap-
sules with cement paste before mixing into concrete. However, this
method may hardly be applied for micro-capsules. Hilloulin et al. [89]
concrete mixing forces and can break when cracks appear. The selected
polymers were brittle at room temperature and have a relatively low
glass transition temperature [89]. Before mixing, these capsules need
to be heated to rubbery state, so that can resist the mixing forces [89].
It seems to be a step forward on the design of capsules for self-healing
in concrete, compared to the method of coating mentioned above. It
still needs great effort to exploring on how to manufacture appropriate
capsules massively for self-healing in concrete.
3.1.4. Efficiency of supply of liquid healing agents by means of capsules
The capsule dosage, size and shape determine the amount of liquid
healing agent released into cracks. It is conceivable that the larger the
capsule dosage, the more liquid healing agent released into cracks. As
indicated in Fig. 9, with the same probability (95%), the efficiency of
the supply of liquid healing agent by spherical capsules increases with
the increase of capsule size. It can be imaged that with the same capsule
dosage, the smaller the capsules the more homogenous the distribution
of capsules is. It means that smaller volume of capsules can be “connect-
ed” with the crack when they are intersected by the crack. Hence, with
the same capsule dosage, the smaller the capsules the lower the effi-
ciency of the supply of liquid healing agent is. Moreover, as
Fig. 9. Amount of liquid healing agent released from spherical capsules with different di-
ameters. The area of the crack is the same as the cross section area of representative vol-
ume element (RVE) of the cement paste whose size is 50 times as large as the capsule
diameter [92].
506 H. Huang et al. / Materials and Design 92 (2016) 499–511
Fig. 10. Normalized liquid supply efficiency versus aspect ratio of cilindrical capsules. Nor-
malized liquid supply efficiency refers to the ratio of the efficiency of the supply of liquid
by cylindrical capsules to that by spherical capsules. Capsule aspect ratio refers to the ratio
of the capsule length to the capsule diameter. For spherical capsule, its aspect ratio is 1.0
(reproduced from [90]).
demonstrated by Mookhoek [90], changing the spherical capsules to
elongated capsules can also enhance the efficiency of the supply of liq-
uid by capsules. As shown in Fig. 10, the ratio of the efficiency of the sup-
ply of liquid by cylindrical capsules to that by spherical capsules, which
is defined as normalized liquid supply efficiency, increases with the in-
crease of capsule aspect ratio (the ratio of the capsule length to the cap-
sule diameter). Moreover, this ratio is independent of the capsule
dosage [90].
3.2. By using vascular systems
For delivering liquid healing agents to cracks a vascular system can
be also applied. As shown in Fig. 11, if cracks intersect any part of the
vascular system inside the material, liquid healing agents can be
transported to the cracks through the vascular system [97,99–102]. As
a result, self-healing of cracks takes place.
Dry [97] embedded some pipes in concrete as a simple vascular sys-
tem to supply healing agents to cracks and found high self-healing ca-
pacity. Stimulated by this positive result, some researchers are now
investigating methods to install vascular systems in concrete structures.
For instance, Pareek et al. [103] set up vascular systems in concrete
Fig. 11. Reinforced concrete frame with a vascular system for self-healing tested by Dry [97]. (Drawing is modified from [9]).
Fig. 12. Porous concrete for vascular systems [104].
structures by creating canal networks. To do that, steel bars with
smooth surface are pre-embedded inside the concrete matrix during
casting. After the concrete is cured for 1 day, the embedded steel bars
are pulled out. As a result, a canal is created inside the concrete matrix.
A network is formed when these canals are connected with each other.
Sangadji and Schlangen [104] are also investigating the installation of a
vascular system in concrete structures. “Porous network concrete” (see
Fig. 12) is developed for creating vascular systems. The pores inside the
porous concrete are of high connectivity and the pore size is in the order
of magnitude of millimeters. Similar to the canal networks proposed by
Pareek et al. [103], when any part of the porous concrete network is
intersected by cracks, liquid healing agents can be delivered (by
means of injection or pumping) to cracks via this system.
As discussed before, one of the disadvantages of the use of capsules is
that only a limited amount of liquid healing agent can be supplied to
cracks. In comparison, sufficient liquid agents can be delivered to cracks
by using vascular systems, leading to high efficiency of self-healing.
However, extra efforts are needed to inject the liquid healing agents
into the vascular systems. Moreover, by now it is still difficult to install
a vascular system inside the concrete matrix [92].
4. Favorable situations in cementitious materials for self-healing
It is well known that self-healing is influenced by material itself and
environmental conditions. In this section, favorable situations in cemen-
titious materials for self-healing will be discussed.
 H. Huang et al. / Materials and Design 92 (2016) 499–511
4.1. Reduction of crack width
It has been demonstrated by many researchers that cracks with a
small width can be healed more easily than those with a large width.
For cracks in engineered cementitious composites (ECC), for instance,
their widths are usually less than 60 μm [105]. Studies on autogenous
self-healing in ECC reveal that ECC has very good self-healing capacity
because the widths of cracks are small [6,29,82,106]. Similar phenom-
ena were observed in other fiber reinforced cementitious materials
[107]. In addition to fibers, shape memory alloys or shrinkable polymers
were used to reduce the crack width in order to increase self-healing ef-
ficiency [98,108–110]. It is conceivable that space of crack with a small
width is smaller than that of crack with a large width. As a result, less re-
action products of self-healing are needed to fill the crack with a small
width.
4.2. Nucleation sites for precipitation provided by fibers
Homma et al. [107] investigated the formation of crystals on the sur-
faces of fibers during the healing process. They found that crystallization
products were attached to polyethylene (PE) fibers easily when the fi-
bers bridged the crack surfaces [107]. In comparison, crystallization
products attached on the surface of steel fibers were much less [107].
Based on the investigation by Homma et al. [107], it can be seen that
in addition to controlling the crack width for self-healing, some fibers
can provide nucleation sites for precipitation of reaction products, lead-
ing to higher efficiency of autogenous self-healing.
4.3. Block of crack pathway with swelling particles
Researchers from Delft University of Technology and Ghent Univer-
sity explored the feasibility of crack-sealing by adding superabsorbent
polymers (SAP) into concrete [111–113]. They demonstrated that
when crack occurs, SAP exposed at crack surfaces swell after absorbing
water and block the crack pathway [111,113]. As a result, permeability
through the cracked sample decreases significantly. Besides, experi-
mental results show that SAP can promote precipitation of minerals in
cracks [111]. Therefore, swelling particles, i.e. SAP facilitate autogenous
self-healing of cracks in cementitious materials.
4.4. High cement content and coarse cement particle
In general, high performance concrete is designed to have higher ce-
ment content. At the same curing age, high performance concrete has
more unhydrated cement than ordinary concrete. As a result, more hy-
dration products can be formed in cracks within the same period of self-
healing [25,114,115]. In addition to the crack filling, hydration of
Table 2
Required conditions for different mechanisms of self-healing.
Mechanisms Conditions
Presence of water in cracks Presence of CO2−
3 ions in
cracks
Not necessary. But the
Autogenous Yes. Water should be continuously
presence of CO2− 3 ions
self-healing available.
facilitates the healing.
Self-healing based Yes. Water should be continuously Not necessary.
on mineral available
admixtures
Self-healing based Yes. Water should be continuously Yes. CO2−
3 ions should be
on bacteria available. continuously available.
Self-healing based Depending on the type of adhesive agents. Not necessary.
on adhesive In most cases cracks should be absent
agents with water.
507
unhydrated cement with higher fraction in high performance concrete
can make the microstructure adjacent to the crack surfaces become
much denser and thus higher capacity of autogenous self-healing [39],
as discussed in Section 2.1.5.
The fraction of unhydrated cement in concrete made of coarse ce-
ment is higher than that in concrete made of fine cement when the mix-
ture is the same. Therefore, concrete made of coarse cement may have
higher potential of autogenous self-healing compared to that made of
fine cement [34].
5. Self-healing mechanisms potentially applied for reinforced con-
crete structures according to service conditions
5.1. Summaries on required conditions for different mechanisms of self-
healing
Different mechanisms of self-healing require different conditions. As
summarized in Table 2, presence of water in cracks is necessary for au-
togenous self-healing and self-healing based on mineral admixtures or
bacteria. For self-healing based on adhesive agents, presence of water
in cracks could have negative effects. Most of the adhesive agents used
for self-healing in cementitious materials cannot get hardened in water.
By now, almost all types of bacteria explored for self-healing in ce-
mentitious materials are the ones that can induce or promote formation
of CaCO3. For self-healing based on bacteria the presence of CO2−
3 ions is
necessary. For autogenous self-healing, the presence of CO2− 3 ions can
enhance the healing efficiency when portlandite in cracks is carbonated.
But the presence of CO2−3 ions in autogenous self-healing is not a prima-
ry condition.
It was reported that by means of autogenous self-healing only the
cracks with a width less than 50 μm can be healed [106]. However, ac-
cording to design standard of reinforced concrete structures, width of
cracks can be up to 200 μm. Mechanisms of self-healing based on min-
eral admixtures, bacteria or adhesive agents can be appreciable options
for such cracks. However, for these three mechanisms of self-healing, it
is recommended to protect the healing agents before the healing agents
are added into concrete. Cost also increases because of the adding of
healing agents while autogenous self-healing is “gift”.
5.2. Service conditions vs. self-healing
Concrete structures serve under various conditions. As presented in
Table 3, some concrete structures serve under water. Some are built for
containing or transporting water. In these structures, cracks can cause
service life and leakage problems. As discussed above, self-healing of
cracks is a vital solution to these problems. As mentioned in
Section 5.1, one of the most necessary conditions for autogenous self-
Crack width Methods to supply healing agent to cracks
Cracks with a width less than 50
–
μm can be healed [105].
Cracks with a width of 200 μm Healing agents, i.e. mineral admixtures,
can be healed [46]. can be added into concrete directly or
encapsulated.
Cracks with a width of 450 μm It is necessary to immobilize bacteria
can be healed [60]. before adding them into concrete.
Wide cracks can be healed It is necessary to supply the healing agent
depending on the amount of to cracks by using capsules or vascular
agents released. systems.
508 H. Huang et al. / Materials and Design 92 (2016) 499–511

Table 3
Serving conditions of reinforced concrete structures and potentially applied self-healing mechanisms.

Structures Mechanism

Autogenous self-healing Self-healing based on Self-healing based on Self-healing based on adhesive agents
mineral bacteria
admixtures

Concrete structures submersed in Recommended Recommended Recommended Hardly

water
Concretea structures under ground Recommended Recommended Recommended For concrete in water-rich regions, it is hardly
applied.
Concrete structures in the open air Need to supply additional Need to supply additional Need to supply additional Recommended
water water water
Indoor concrete structures Hardly Hardly Hardly Recommended
elements
a
For concrete underground in water-rich region, here it is assumed that cracks exposed to material surfaces contacting to soil can be saturated easily with water.

healing and self-healing based on bacteria or mineral admixtures is that
water is present in cracks (see Table 2). Therefore, for concrete struc-
tures under water, autogenous self-healing and self-healing based on
bacteria or mineral admixtures can be adopted, as shown schematically
in Fig. 13. Self-healing based on adhesive agents is not suitable for such
concrete structures because the presence of water in cracks may inhibit
the release of adhesive agents into cracks and the hardening of adhesive
agents (see Fig. 13).
For underground concrete structures, such as storage and garage in
water-rich regions, water can also penetrate into cracks that are ex-
posed to soil. This is similar to the case that concrete structures are
under water. As shown schematically in Fig. 13, autogenous self-
healing and self-healing based on bacteria or mineral admixtures are
recommended, rather than self-healing based on adhesive agents. How-
ever, in some regions water is not extremely rich. It takes some time for
water to fill the cracks. For this case self-healing based on adhesive
agents is a possible mechanism to be applied. Moreover, wetting-
drying cycles usually take place in cracks of concrete structures (see
Fig. 13). Wetting-drying cycles facilitate penetration of CO2 into cracks
when some part of the crack is exposed to air (for example, cracks prop-
agating through the wall of tunnel). As shown in Table 2, dissolved CO2
facilitate the precipitation of calcite during autogenous healing and the
healing based on bacteria.
For concrete structures in the open air, liquid water is absent in
cracks. However, cracks provide pathway for CO2 to access to the inter-
face between reinforcement bars and concrete matrix. Carbonation of
concrete surrounding reinforcement bars increases risk of corrosion of
reinforcement bars when chloride ions are present in the moisture
(for example, in offshore environment). As a result, service life of con-
crete structures is shortened. As discussed before, self-healing of cracks
can solve this problem. However, the lack of water in cracks makes au-
togenous self-healing and self-healing based on bacteria or mineral ad-
mixtures very difficult. Additional water should be supplied to cracks for
such self-healing techniques. The supply of additional water can be

Fig. 13. Schematic of concretes under different service conditions and corresponding suitable mechanisms of self-healing.
 H. Huang et al. / Materials and Design 92 (2016) 499–511
done by using a vascular system described in Section 3. In comparison,
self-healing based on adhesive agents may be more convenient under
this condition (concrete structures serve in the open air).
For indoor concrete elements in structures, it is difficult to supply ad-
ditional water to cracks via a vascular system (water can easily leaks out
from cracks and the functionality of structures are negatively affected).
In this case, autogenous self-healing and self-healing based on bacteria
or mineral admixtures are hardly applied. Self-healing based on adhe-
sive agents may be most suitable for this particular service condition.
6. “Profit” from self-healing in cementitious materials
Self-healing in cementitious materials has been investigated for de-
cades. However, self-healing has not yet been successfully applied in
concrete structures so far. The uncertainty about the profit from self-
healing is the main reason. In this subsection, the additional costs for ap-
plying self-healing techniques in engineering practice and the potential
gains from self-healing are discussed.
6.1. Additional cost for applying self-healing in cementitious materials
As mentioned before, there is no additional cost for applying autog-
enous self-healing in cementitious materials. It is an inherent perfor-
mance of cementitious materials, although it proceeds slowly and is
only able to heal small cracks (usually less than 50 μm). In order to im-
prove the capacity of self-healing, healing agents added into mixture of
concrete, such as bacteria, mineral admixtures and adhesive agents, are
necessary. Indeed, the cost of concrete with such healing agents added
increases.
According to Jonkers [116], the price of self-healing concrete based
on bacteria is about twice that of traditional concrete. It will decrease
as this technique is being improved. As demonstrated by Jonkers
[116], the price of self-healing concrete based on bacteria can reduce
dramatically if researchers can get the bacteria to use a sugar-based nu-
trient in the future.
For self-healing based on mineral admixtures, because the prices of
mineral admixtures are close to that of cement, the additional cost
will not be high when they are used to replace some of the cement in
concrete. If they are coated before added into concrete, the cost will in-
crease and it depends on the coating technique.
For self-healing based on adhesive agents, the additional cost should
be much higher than that self-healing based on mineral admixtures. On
one hand, the prices of adhesive agents are usually high. On the other
hand, the encapsulation of adhesive agents increases the cost dramati-
cally. To encapsulate liquid-based adhesive agents, chemical synthesis
techniques are usually applied [88,89]. Their costs are much higher
than the cost for coating mineral admixtures.
Based on the discussion above, it is found that the additional costs
for self-healing are still difficult to quantify because some key tech-
niques for realizing self-healing, such as encapsulation of adhesive
agents and coating of mineral admixtures, are still being developed.
6.2. Potential gains from self-healing in cementitious materials
It has been confirmed that the penetration of aggressive ions into
concrete through cracks decreases significantly after self-healing
[117–119]. Therefore, corrosion of reinforcement bars and degradation
of concrete can be inhibited. According to literature survey, it is agreed
well that the direct benefits of self-healing in concrete are the reduction
of maintenance budget for concrete structures and the extension of ser-
vice life of concrete structures. Moreover, because of the reductions of
maintenances and reconstruction, CO2 emission decreases dramatically.
Absolutely, self-healing of cracks has significant beneficial on economy,
society and environment.
However, by now most of the investigations on self-healing have
been carried out in laboratory. The studies of self-healing in-situ are
509
Fig. 14. Schematic of repeatable self-healing concept and the extension of service life (im-
proved based on [2]).
much far from satisfactory. The repeatability of self-healing of a
crack under repeated loads (as schematically shown in Fig. 14) are
still unknown. The quantitative relationship between self-healing
of cracks and extension of service life of concrete structures (see
Fig. 14) is not yet determined. The lack of the aforementioned infor-
mation limits the applications of self-healing concepts in concrete
structures.
7. Summaries and perspectives
In this paper, mechanisms of self-healing, i.e. autogenous self-
healing, self-healing based on mineral admixtures, self-healing based
on bacteria and self-healing based on adhesive agents, are reviewed. Lit-
erature shows that all mechanisms of self-healing are effective, to some
extent, under some particular conditions.
Based on the previous studies, favorable situations of cementitious
materials for self-healing are summarized. They includes small crack
width, high cement content, the use of coarse cement and nucleation
sites for precipitation provided by fibers etc.
Required environmental conditions for each mechanism of self-
healing are analyzed. The presence of water in cracks is required for au-
togenous self-healing and self-healing based on mineral admixtures or
bacteria, while for self-healing based on adhesive agents the presence
of water in cracks could have negative effects. For self-healing based
on bacteria the presence of CO2− 3 ions is also a primary condition, but
not for other mechanisms of self-healing. Therefore, for real engineering
applications self-healing techniques should be chosen according to the
service conditions of concrete structures.
Additional costs for applying self-healing in concrete structures and
the benefits from self-healing are also discussed in this paper. It is found
that the additional costs for applying self-healing in engineering prac-
tice are still difficult to quantify because some key techniques for realiz-
ing self-healing, such as encapsulation of adhesive agents and coating of
mineral admixtures, are still being developed.
It has to be admitted that by now most of the investigations on self-
healing have been carried out in laboratory. The studies of self-healing
in-situ are much far from satisfactory. The repeatability of self-healing
of a crack under repeated loads are still unknown. The quantitative rela-
tionship between self-healing of cracks and extension of service life of
concrete structures is not yet determined. The lack of the aforemen-
tioned information limits the applications of self-healing techniques in
concrete construction. Intensive investigations on these aspects are
emergent.
510 H. Huang et al. / Materials and Design 92 (2016) 499–511
Acknowledgments
The authors would like to thank the Nature Science Fund of Jiangsu
Province in China (BK20150601), National Basic Research Program of
China (973 Program: 2011CB013800) and National Nature Science
Fund of China (51178104) for the financial support.
References
[1] K. van Breugel, Self-healing material concepts as solution for aging infrastructure,
37th Conference on Our World in Concrete & Structures, 2012 (Singapore).
[2] K. van Breugel, Is there a market for self-healing cement-based materials? 1st In-
ternational Conference on Self-healing Materials, Noordwijk aan zee, the
Netherlands, 2007.
[3] E. Cailleux, V. Pollet, Investigations on the development of self-healing properties
in protective coatings for concrete and repair mortars, 2nd International Confer-
ence on Self Healing Materials, 2009 (Chicago).
[4] K. van Tittelboom, Self-healing concrete through incorporation of encapsulated
bacteria- or polymer-based healing agents, Ghent University, Ghent, 2012 344.
[5] L. Ferrara, V. Krelani, M. Carsana, A “fracture testing” based approach to assess
crack healing of concrete with and without crystalline admixtures, Constr. Build.
Mater. 68 (0) (2014) 535–551.
[6] V.C. Li, E. Herbert, Robust self-healing concrete for sustainable infrastructure, J.
Adv. Concr. Technol. 10 (2012) 207–218.
[7] M. Wu, B.R. Johannesson, M. Geiker, A review: self-healing in cementitious mate-
rials and engineered cementitious composite as a self-healing material, Constr.
Build. Mater. 28 (1) (2012) 571–583.
[8] K. Van Tittelboom, N. De Belie, Self-healing in cementitious materials—a review,
Materials 6 (6) (2013) 2182–2217.
[9] H. Mihashi, T. Nishiwaki, Development of engineered self-healing and self-
repairing concrete, J. Adv. Concr. Technol. 10 (2012) 170–184.
[10] M. de Rooij, K. van Tittelboom, N. De Belie, E. Schlangen, Self-Healing Phenomena
in Cement-Based Materials: State-of-the-Art Report of RILEM Technical Committee
221-SHC: Self-Healing Phenomena in Cement-Based Materials, Springer, 2013.
[11] N. Hearn, Self-sealing, autogenous healing and continued hydration: what is the
difference? Mater. Struct. 31 (8) (1998) 563–567.
[12] G.W. Hyde, W.J. Smith, Results of experiments made to determine the permeability
of cements and cement mortars, J. of the Franklin Institute Philadelphia (1889)
199–207.
[13] W.H. Glanville, The permeability of Portland cement concrete, Building Research,
Technical Paper 3 (1931) 1–61.
[14] V.J. Soroker, A.J. Denson, Autogenous healing of concrete, Zement 25 (30) (1926).
[15] F. Brandeis, Autogenous healing of concrete, Beton u Eisen 36 (12) (1937).
[16] C.A. Clear, The effect of autogenous healing upon leakage of water through cracks
in concrete, Tech. Report 559 (1985).
[17] N. Hearn, Saturated Permeability of Concrete as Influenced by Cracking and Self-
sealing, University of Cambridge, 1992.
[18] C. Edvardsen, Water permeability and autogenous healing of cracks in concrete,
ACI Mater. J. 96 (4) (1999) 448–454.
[19] H.-W. Reinhardt, M. Jooss, Permeability and self-healing of cracked concrete as a
function of temperature and crack width, Cem. Concr. Res. 33 (7) (2003) 981–985.
[20] P. Fidjestol, N. Nilsen, Field test of reinforcement corrosion in concrete, ACI Special
Publications (1980) 205–221.
[21] W. Ramm, M. Biscoping, Autogenous healing and reinforcement corrosion of
water-penetrated separation cracks in reinforced concrete, Nucl. Eng. Des. 179
(2) (1998) 191–200.
[22] N. Otsuki, S. Miyazato, N.B. Diola, H. Suzuki, Influences of bending crack and water-
cement ratio on chloride-induced corrosion of main reinforcing bars and stirrups,
ACI Mater. J. 97 (4) (2000) 454–464.
[23] K.R. Lauer, F.O. Slate, Autogenous healing of cement paste, ACI Mater. J. 52 (1956).
[24] R.K. Dhir, C.M. Sangha, J.G.L. Munday, Strength and deformation properties of au-
togenously healed mortars, ACI Mater. J. 70 (3) (1973) 231–236.
[25] S. Granger, A. Loukili, G. Pijaudier-Cabot, G. Chanvillard, Experimental characteriza-
tion of the self-healing of cracks in an ultra high performance cementitious mate-
rial: mechanical tests and acoustic emission analysis, Cem. Concr. Res. 37 (4)
(2007) 519–527.
[26] S. Jacobsen, J. Marchand, H. Hornain, Sem observations of the microstructure of
frost deteriorated and self-healed concretes, Cem. Concr. Res. 25 (8) (1995)
1781–1790.
[27] E. Schlangen, N. ter Heide, K. van Breugel, in: M.S. Konsta-Gdoutos (Ed.), Crack
Healing of Early Age Cracks in Concrete, in Measuring, Monitoring and Modeling
Concrete Properties, Springer, Netherlands, 2006.
[28] Y. Yang, M.D. Lepech, E.-H. Yang, V.C. Li, Autogenous healing of engineered cemen-
titious composites under wet-dry cycles, Cem. Concr. Res. 39 (5) (2009) 382–390.
[29] S. Qian, J. Zhou, M.R. de Rooij, E. Schlangen, G. Ye, K. van Breugel, Self-healing be-
havior of strain hardening cementitious composites incorporating local waste ma-
terials, Cem. Concr. Compos. 31 (9) (2009) 613–621.
[30] K. Sisomphon, O. Copuroglu, E.A.B. Koenders, Self-healing of surface cracks in mor-
tars with expansive additive and crystalline additive, Cem. Concr. Compos. 34 (4)
(2012) 566–574.
[31] J. Parks, M. Edwards, P. Vikesland, A. Dudi, Effects of bulk water chemistry on au-
togenous healing of concrete, J. Mater. Civ. Eng. 22 (5) (2010) 515–524.
[32] H. Huang, G. Ye, D. Damidot, Characterization and quantification of self-healing be-
haviors of microcracks due to further hydration in cement paste, Cem. Concr. Res.
52 (2013) 71–81.
[33] B. Bary, A. Sellier, Coupled moisture—carbon dioxide–calcium transfer model for
carbonation of concrete, Cem. Concr. Res. 34 (10) (2004) 1859–1872.
[34] K. van Breugel, Caring for ageing infrastructure — scope, strategy and responsible
stewardship, 3rd International Conference on Service Life Design for Infrastruc-
tures, 2014 (Zhuhai).
[35] K. Van Tittelboom, E. Gruyaert, H. Rahier, N. De Belie, Influence of mix composition
on the extent of autogenous crack healing by continued hydration or calcium car-
bonate formation, Constr. Build. Mater. 37 (0) (2012) 349–359.
[36] P. Termkhajornkit, T. Nawa, Y. Yamashiro, T. Saito, Self-healing ability of fly ash–
cement systems, Cem. Concr. Compos. 31 (3) (2009) 195–203.
[37] M. Sahmaran, G. Yildirim, T.K. Erdem, Self-healing capability of cementitious com-
posites incorporating different supplementary cementitious materials, Cem. Concr.
Compos. 35 (1) (2013) 89–101.
[38] H. Huang, G. Ye, D. Damidot, Effect of blast furnace slag on self-healing of
microcracks in cementitious materials, Cem. Concr. Res. 60 (0) (2014) 68–82.
[39] H. Huang, Thermodynamics of Autogenous Self-healing in Cementitious Materials,
Delft University of Technology, 2014.
[40] J. van der Lee, L.C.H.E.S.S. de Windt, 1999 Available from: http://chess.geosciences.
ensmp.fr/.
[41] H. Huang, G. Ye, Self-healing of cracks in cement paste affected by additional Ca2+
ions of healing agent, J. Intell. Mater. Syst. Struct. 26 (3) (2015) 309–320.
[42] D. Palin, V. Wiktor, H.M. Jonkers, Autogenous healing of marine exposed concrete:
characterization and quantification through visual crack closure, Cem. Concr. Res.
73 (0) (2015) 17–24.
[43] H. Huang, G. Ye, L. Pel, New insights into autogenous self-healing in cement paste
based on nuclear magnetic resonance (NMR) tests, Mater. Struct. (2015)http://dx.
doi.org/10.1617/s11527-015-0664-9 (online).
[44] J. Liu, F. Xing, B. Dong, H. Ma, D. Pan, Study on water sorptivity of the surface layer
of concrete, Mater. Struct. (2014) 47(11).
[45] S. Nagataki, H. Gomi, Expansive admixtures (mainly ettringite), Cem. Concr.
Compos. 20 (2–3) (1998) 163–170.
[46] T. Kishi, T.H. Ahn, A. Hosoda, S. Suzuki, H. Takaoka, Self-healing behaviour by ce-
mentitious recrystallization of cracked concrete incorporating expansive agent,
1st International Conference on Self-healing Materials, 2007 (Dordrecht).
[47] T.H. Ahn, T. Kishi, Crack self-healing behavior of cementitious composites incorpo-
rating various mineral admixtures, J. Adv. Concr. Technol. 8 (2) (2010) 16.
[48] T. Kishi, T.H. Ahn, M. Morita, T. Koide, Field test of self-healing concrete on the re-
covery of water tightness to leakage through cracks, The 3rd International Confer-
ence on Self-healing Materials, 2011 (Bath, UK).
[49] M. Roig-Flores, S. Moscato, P. Serna, L. Ferrara, Self-healing capability of concrete
with crystalline admixtures in different environments, Constr. Build. Mater. 86
(0) (2015) 1–11.
[50] Z. Jiang, W. Li, Z. Yuan, Influence of mineral additives and environmental condi-
tions on the self-healing capabilities of cementitious materials, Cem. Concr.
Compos. 57 (0) (2015) 116–127.
[51] G. Yildirim, M. Sahmaran, M. Balcikanli, E. Ozbay, M. Lachemi, Influence of cracking
and healing on the gas permeability of cementitious composites, Constr. Build.
Mater. 85 (0) (2015) 217–226.
[52] R. Sahamitmongkol, T. Kishi, Tensile behavior of restrained expansive mortar and
concrete, Cem. Concr. Compos. 33 (1) (2011) 131–141.
[53] T. Kishi, T.H. Ahn, T. Koide, Crack self-healing concrete introducing semicapsulation
effect by granulation of ingredients, Proceedings of the 3rd JCI-KCI-TCI Symposium
on Recent Advancement of Technologies in Concrete Engineering, Japan Concrete
Institute, Hiroshima, 2012.
[54] H. Dong, H. Huang, G. Ye, Inorganic powder encapsulated in brittle polymer parti-
cles for self-healing cement-based materials, 4th International Conference on Self-
healing Materials, 2013 (Ghent).
[55] Y.-S. Lee, J.-S. Ryou, Self healing behavior for crack closing of expansive agent
via granulation/film coating method, Constr. Build. Mater. 71 (0) (2014)
188–193.
[56] U.K. Gollapudi, C.L. Knutson, S.S. Bang, M.R. Islam, A new method for controlling
leaching through permeable channels, Chemosphere 30 (4) (1995) 695–705.
[57] W. De Muynck, N. De Belie, W. Verstraete, Microbial carbonate precipitation in
construction materials: a review, Ecol. Eng. 36 (2) (2010) 118–136.
[58] M. Luo, C.-X. Qian, R.-Y. Li, Factors affecting crack repairing capacity of bacteria-
based self-healing concrete, Constr. Build. Mater. 87 (0) (2015) 1–7.
[59] H. Jonkers, Self healing concrete: a biological approach, in: S. Zwaag (Ed.), Self
Healing Materials, Springer, Netherlands 2008, pp. 195–204.
[60] H.M. Jonkers, A. Thijssen, G. Muyzer, O. Copuroglu, E. Schlangen, Application of
bacteria as self-healing agent for the development of sustainable concrete, Ecol.
Eng. 36 (2) (2010) 230–235.
[61] V. Wiktor, H.M. Jonkers, Quantification of crack-healing in novel bacteria-based
self-healing concrete, Cem. Concr. Compos. 33 (7) (2011) 763–770.
[62] H. Jonkers, E. Schlangen, Towards a sustainable bacterially-mediated self-
healing concrete, 2nd International Conference on Self Healing Materials,
2009 (Chicago).
[63] C. Stuckrath, R. Serpell, L.M. Valenzuela, M. Lopez, Quantification of chemical and
biological calcium carbonate precipitation: performance of self-healing in rein-
forced mortar containing chemical admixtures, Cem. Concr. Compos. 50 (0)
(2014) 10–15.
[64] J. Wang, K. Van Tittelboom, N. De Belie, W. Verstraete, Use of silica gel or polyure-
thane immobilized bacteria for self-healing concrete, Constr. Build. Mater. 26 (1)
(2012) 532–540.
 H. Huang et al. / Materials and Design 92 (2016) 499–511
[65] R. Wang, C. Qian, J. Wang, Bio-deposition of a calcite layer on cement-based mate-
rials by brushing with agar-immobilised bacteria, Adv. Cem. Res. 23 (4) (2011)
185–192.
[66] C. Qian, M. Luo, Q. Pan, R. Li, Mechanism of formation of calcite in microbe-based
self-healing concrete (in Chinese), J. Chin. Ceram. Soc. 41 (5) (2013) 620–626.
[67] C. Qian, R. Li, Q. Pan, M. Luo, H. Rong, Efficiency of self-healing in concrete based on
microbe (in Chinese), Journal of Southeast University 43 (2) (2013) 360–364.
[68] L. Ren, C. Qian, Restoration of cracks on surface of cement-based materials by car-
bonic anhydrase microbiologically precipitation calcium carbonate, J. Chin. Ceram.
Soc. 42 (11) (2014) 1389–1395.
[69] K. Van Tittelboom, N. De Belie, D. Van Loo, P. Jacobs, Self-healing efficiency of ce-
mentitious materials containing tubular capsules filled with healing agent, Cem.
Concr. Compos. 33 (4) (2011) 497–505.
[70] H. Mihashi, Y. Kaneko, Fundamental study on development of intelligent concrete
with self-healing capability for prevention of water leakage, Journal of Architecture
and Building Science 115 (1456) (2000) 1–4.
[71] F. Xing, Z. Ni, Self-healing mechanism of a novel cementitious composite using mi-
crocapsules, International Conference on Durability of Concrete Structres: Ad-
vances in Conference on Durability of Concrete Structures, 2008 (Hangzhou,
China).
[72] I. Kaltzakorta, E. Erkizia, Silica microcapsules encapsulating epoxy compounds for
self-healing cementitious materials, The 3rd International Conference on Self-
healing Materials, 2011 (Bath, UK).
[73] T.D.P. Thao, T.J.S. Johnson, Q.S. Tong, P.S. Dai, Implementation of self-healing in
concrete — proof of concept, The IES Journal Part A: Civil & Structural Engineering
2 (2) (2009) 116–125.
[74] T. Nishiwaki, H. Mihashi, B.K. Jang, K. Miura, Development of self-healing system
for concrete with selective heating around crack, J. Adv. Concr. Technol. 4 (2)
(2006) 267–275.
[75] Z. Yang, J. Hollar, X. He, X. Shi, A self-healing cementitious composite using oil core/
silica gel shell microcapsules, Cem. Concr. Compos. 33 (4) (2011) 506–512.
[76] C. Dry, W. McMillan, Three-part methylmethacrylate adhesive system as an inter-
nal delivery system for smart responsive concrete, Smart Mater. Struct. 5 (3)
(1996) 297–300.
[77] C. Joseph, A. Jefferson, B. Isaacs, R. Lark, D. Gardner, Experimental investigation of
adhesive-based self-healing of cementitious materials, Mag. Concr. Res. 62 (11)
(2010) 831–843.
[78] C.M. Dry, Three designs for the internal release of sealants, adhesives, and water-
proofing chemicals into concrete to reduce permeability, Cem. Concr. Res. 30
(12) (2000) 1969–1977.
[79] C. Dry, Repair and preventation of damage due to transverse shrinkage cracks in
bridge decks, Symposium on Smart Structures and Materials, 1999.
[80] C. Joseph, A. Jefferson, M. Canoni, Issues relating to the autonomic healing of ce-
mentitious materials, 1st International Conference on Self Healing Materials
2007, p. 53 (Noordwijk aan Zee, The Netherlands).
[81] C. Joseph, Experimental and Numerical Study of the Fracture and Self-healing of
Cementitious Materials, Cardiff University, 2008.
[82] V.C. Li, Y.M. Lim, Y.-W. Chan, Feasibility study of a passive smart self-healing ce-
mentitious composite, Compos. Part B 29 (6) (1998) 819–827.
[83] L. Sun, W. Yu, Q. Ge, Experimental research on the self-healing performance of
micro-cracks in concrete bridge, Adv. Mater. Res. 250-253 (2011) 28–32.
[84] C. Joseph, A. Jefferson, B. Isaacs, R. Lark, D. Gardner, Adhesive-based self-healing of
cementitious materials, 2nd International Conference on Self Healing Materials
2009, p. 121 (Chicago, USA).
[85] C. Dry, M. Corsaw, E. Bayer, A comparison of internal self-repair with resin injection
in repair of concrete, J. Adhes. Sci. Technol. 17 (1) (2003) 79–89.
[86] S.R. White, N.R. Sottos, P.H. Geubelle, J.S. Moore, M.R. Kessler, S.R. Sriram, E.N.
Brown, S. Viswanathan, Autonomic healing of polymer composites, Nature 409
(6822) (2001) 794–797.
[87] J.R. Sara, Encapsulation: overview of uses and techniques, Encapsulation and Con-
trolled Release of Food Ingredients, American Chemical Society 1995, pp. 2–7.
[88] B. Dong, Y. Wang, G. Fang, N. Han, F. Xing, Y. Lu, Smart releasing behavior of a
chemical self-healing microcapsule in the stimulated concrete pore solution,
Cem. Concr. Compos. 56 (0) (2015) 46–50.
[89] B. Hilloulin, K. Van Tittelboom, E. Gruyaert, N. De Belie, A. Loukili, Design of poly-
meric capsules for self-healing concrete, Cem. Concr. Compos. 55 (0) (2015)
298–307.
[90] S.D. Mookhoek, Novel Routes to Liquid-based Self-healing Polymer Systems, Delft
University of Technology, 2010.
[91] S.V. Zemskov, H.M. Jonkers, F.J. Vermolen, Two analytical models for the probabil-
ity characteristics of a crack hitting encapsulated particles: application to self-
healing materials, Comput. Mater. Sci. 50 (12) (2011) 3323–3333.
511
[92] H. Huang, G. Ye, Z. Shui, Feasibility of self-healing in cementitious materials — by
using capsules or a vascular system? Constr. Build. Mater. 63 (0) (2014) 108–118.
[93] Z. Lv, H. Chen, H. Yuan, Analytical solution on dosage of self-healing agents in ce-
mentitious materials: long capsule model, J. Intell. Mater. Syst. Struct. (2012) 1–12.
[94] S.D. Mookhoek, H.R. Fischer, S.V.D. Zwaag, A numerical study into the effects of
elongated capsules on the healing efficiency of liquid-based systems, Comput.
Mater. Sci. 47 (2) (2009) 506–511.
[95] A. Papoulis, Probability, Random Variables and Stochastic Processes, third ed.
Mcgraw-Hill College, 1991.
[96] F.A. Gilabert, D. Garoz, W. Van Paepegem, Stress concentrations and bonding
strength in encapsulation-based self-healing materials, Mater. Des. 67 (0) (2015)
28–41.
[97] C. Dry, Design of self-growing, self-sensing and self-repairing materials for engi-
neering applications, Proceedings of SPIE 2001, pp. 23–29.
[98] Y. Kuang, J. Ou, Self-repairing performance of concrete beams strengthened using
superelastic SMA wires in combination with adhesives released from hollow fibers,
Smart Mater. Struct. 17 (2) (2008) 1–7.
[99] K.S. Toohey, N.R. Sottos, J.A. Lewis, J.S. Moore, S.R. White, Self-healing materials
with microvascular networks, Nat. Mater. 6 (8) (2007) 581–585.
[100] A.R. Hamilton, N.R. Sottos, S.R. White, Self-healing of internal damage in synthetic
vascular materials, Adv. Mater. 22 (45) (2010) 5159–5163.
[101] A.R. Hamilton, N.R. Sottos, S.R. White, Mitigation of fatigue damage in self-healing
vascular materials, Polymer 53 (24) (2012) 5575–5581.
[102] J.F. Patrick, N.R. Sottos, S.R. White, Microvascular based self-healing polymeric
foam, Polymer 53 (19) (2012) 4231–4240.
[103] S. Pareek, A. Oohira, A fundamental study on regain of flexural strength of mortars
by using a self-repair network system, The 3rd International Conference on Self-
healing Materials, 2011 (Bath, UK).
[104] S. Sangadji, E. Schlangen, Self-healing of concrete structures — novel approach
using porous network concrete, J. Adv. Concr. Technol. 10 (5) (2012) 185–194.
[105] V. Li, S. Wang, C. Wu, Tensile strain-hardening behavior of polyvinyl alcohol
engineered cementitious composite (PVA-ECC), ACI Mater. J. 98 (6) (2011)
483–492.
[106] V.C. Li, E.-H. Yang, Self-healing in concrete materials, in: S. van der Zwaag (Ed.),
Self Healing Materials An Alternative Approach to 20 Centuries of Materials Sci-
ence, Springer, Dordrecht, The Netherlands, 2007.
[107] D. Homma, H. Mihashi, T. Nishiwaki, Self-healing capability of fibre reinforced ce-
mentitious composites, J. Adv. Concr. Technol. 7 (2) (2009) 217–228.
[108] Choi, E., Cho, S.-C., Hu, J.W., Park, T., and Chung, Y.-S., Recovery and residual stress
of SMA wires and applications for concrete structures. Smart Mater. Struct. 19(9):
p. 094013.
[109] A. Jefferson, C. Joseph, R. Lark, B. Isaacs, S. Dunn, B. Weager, A new system for crack
closure of cementitious materials using shrinkable polymers, Cem. Concr. Res. 40
(5) (2010) 795–801.
[110] S. Pareek, K.C. Shrestha, Y. Suzuki, T. Omori, R. Kainuma, Y. Araki, Feasibility of ex-
ternally activated self-repairing concrete with epoxy injection network and Cu–Al–
Mn superelastic alloy reinforcing bars, Smart Mater. Struct. 23 (10) (2014).
[111] A. Mignon, D. Snoeck, D. Schaubroeck, N. Luickx, P. Dubruel, S. Van Vlierberghe, N.
De Belie, pH-responsive superabsorbent polymers: a pathway to self-healing of
mortar, React. Funct. Polym. 93 (0) (2015) 68–76.
[112] D. Snoeck, K. Van Tittelboom, S. Steuperaert, P. Dubruel, N. De Belie, Self-healing
cementitious materials by the combination of microfibres and superabsorbent
polymers, J. Intell. Mater. Syst. Struct. 25 (1) (2014) 13–24.
[113] J.S. Kim, E. Schlangen, Super absorbent polymers to simulate self healing in ECC, in:
K. Van Breugel, G. Ye, Y. Yuan (Eds.),2nd International Symposium on Service Life
Design for Infrastructures, 2010 (Delft, the Netherlands).
[114] D. Fukuda, Y. Nara, Y. Kobayashi, M. Maruyama, M. Koketsu, D. Hayashi, H. Ogawa,
K. Kaneko, Investigation of self-sealing in high-strength and ultra-low-
permeability concrete in water using micro-focus X-ray CT, Cem. Concr. Res. 42
(11) (2012) 1494–1500.
[115] N.T. Heide, Crack Healing in Hydrating Concrete, Delft University of Technology,
2005.
[116] Davis, J.L. Self-Healing Concrete Repairs Its Own Cracks. 2015; Available from:
http://www.iflscience.com/chemistry/self-healing-concrete-repairs-its-own-
cracks.
[117] M. Şahmaran, Effect of flexure induced transverse crack and self-healing on chlo-
ride diffusivity of reinforced mortar, J. Mater. Sci. 42 (22) (2007) 9131–9136.
[118] M. Li, V.C. Li, Cracking and healing of engineered cementitious composites under
chloride environment, ACI Journal Proceedings 108 (3) (2011) 8.
[119] M. Maes, K. Van Tittelboom, N. De Belie, The efficiency of self-healing cementitious
materials by means of encapsulated polyurethane in chloride containing environ-
ments, Constr. Build. Mater. 71 (0) (2014) 528–537.