materials

Article
Self-Healing Concrete Using Rubber Particles to
Immobilize Bacterial Spores
Hongyin Xu, Jijian Lian, Maomao Gao, Dengfeng Fu and Yue Yan *
State Key Laboratory of Hydraulic Engineering Simulation and Safety, Tianjin University, Tianjin 300350, China
* Correspondence: yue.yan@tju.edu.cn; Tel.: +86-22-27404207




Received: 16 May 2019; Accepted: 15 July 2019; Published: 19 July 2019


Abstract: Bacteria-based self-healing concrete is a construction material used to repair cracks in

concrete, in which the bacterial spores are immobilized by bacteria carriers. However, the currently
available bacteria carriers are not always suitable due to a complicated procedure or high cost.
To develop a more suitable bacteria carrier as well as improve the anti-crack capability of self-healing
concrete, in this study we evaluate the feasibility of using rubber particles as a novel bacteria carrier
in self-healing concrete. Two types of self-healing concrete are prepared with rubber particles of
different sizes to quantify the crack-healing effect. In addition, the fluidity and mechanical properties
of the self-healing rubber concrete are compared with those of plain concrete and normal rubber
concrete. The experimental results show that the self-healing rubber concrete with a particle size
of 1~3 mm has a better healing capacity than the self-healing rubber concrete with a particle size of
0.2~0.4 mm, and the width value of the completely healed crack is 0.86 mm. The self-healing rubber
concrete has a higher slump than the plain concrete and normal rubber concrete. According to the
strength tests, the compressive strengths of the self-healing rubber concrete are low early on but they
exceed those of the corresponding normal rubber concrete at 28 days. Moreover, the self-healing
rubber concrete has higher splitting tensile strengths than the plain concrete and a better anti-crack
capability. The results of a comparison to the other two representative bacterial carriers indicate
that rubber particles have potential to be a widely used bacteria carrier for practical engineering
applications in self-healing concrete.

Keywords: self-healing concrete; rubber particles; cracks; strength

1. Introduction
Concrete is a widely used construction material as it has a high compressive strength and good
durability. However, a major concern with concrete is that it is vulnerable to cracking under the
influence of temperature fluctuations and external load due to its low tensile strength and poor
deformation ability [1,2]. The cracks in concrete provide a path for water and other aggressive
substances, such as sulfate and chloride ions, to penetrate into the concrete matrix, resulting in steel
reinforcement corrosion, strength reduction, and structural failure [3–5]. Various methods have been
developed to heal the cracks in concrete. They can be divided into two categories: man-made repairs
and self-healing repairs. Man-made methods apply repairing agents from outside the cracks to
fill them after they are detected, and are mainly used to repair large cracks and difficult to repair
small and deep cracks [5,6]. Self-healing approaches pre-bury the repairing agents in the concrete
matrix and can repair cracks automatically once they appear, which can save on operation costs
compared to the man-made methods [7]. According to previous studies [8–11], the self-healing
approaches include autogenous, adhesive-based, mineral-admixtures-based, and bacteria-based
methods. The bacteria-based self-healing repair method has attracted much attention due to its better
healing capacity and environmentally friendly character.

Materials 2019, 12, 2313; doi:10.3390/ma12142313 www.mdpi.com/journal/materials

Materials 2019, 12, 2313 2 of 16

The bacteria-based self-healing of concrete cracks is based on Microbial Induced Calcite Precipitate
(MICP), which involves the application of mineral-producing bacteria to precipitate calcium carbonate
onto the crack face in the concrete [12,13]. The bacterial spores and other repairing agents are mixed
into the fresh concrete when manufacturing self-healing concrete. They remain dormant in the concrete
matrix and have the potential to germinate before cracks appear. When the concrete cracks, the bacteria
begin to proliferate after activation by the ingress of water and oxygen via the cracks and produce
calcium carbonate on the crack faces. Thus, the generated crystals can block the cracks and result in
the self-repair of the cracks [1,6].
Due to the harsh environment inside the concrete, the bacterial spores must be protected from
squeezing and high alkalinity by a bacteria carrier to ensure that the bacteria have space to grow and a
mineralized environment [14]. A range of bacteria carriers have been adopted in previous studies and
they can be divided into two categories depending on the immobilization mode: encapsulated and
adsorbed (see Table 1). The encapsulated mode injects bacterial spore suspension and other repair
agents into microcapsules made of special materials, such as glass [6], poly [15], and melamine [16].
These microcapsules are embedded in the concrete matrix and rupture as soon as cracks appear.
Thus, the healing agent is released onto the crack face and crystals are produced to heal the cracks [17,18].
Self-healing by the encapsulation method can provide high-quality healing in terms of being able to
repair cracks with a wider range of widths [2,8]. However, the encapsulated method always involves a
complicated procedure and professional equipment is required to produce microcapsules, resulting in
a relatively high cost. In the adsorbed method, bacterial spores are immobilized by porous materials
with high porosity and a high capacity for water absorption. Their highly porous structure can provide
sufficient oxygen, water, and growth space to bacteria after the concrete cracks. Various attempts have
been made to develop porous bacteria carriers and the results vary across different studies [1,7,19,20].
To date, an ideal bacteria carrier has not been obtained. Thus, it is necessary to develop more carrier
techniques for bacteria-based self-healing concrete.

Table 1. The bacteria carriers adopted in the published studies.

Immobilization Mode Bacteria Carrier Immobilization Method Maximum Crack Width (mm)
0.3
Epoxy [17] • Making microcapsules;
Encapsulated • Encapsulating bacterial spores and 0.4
Poly [15] other healing agents;
• Filling the protection materials 0.97
Melamine [16] into the microcapsules.

• Mixing bacterial spores with

polymer solution and an initiator;
Hydrogel [18] • Degassing and subjecting to 0.5
UV irradiation.
• Freezing and drying.

Diatomaceous earth [19] • Impregnating the carriers with a 0.17

Expanded perlite [7] bacterial spore suspension; 0.79
Adsorbed
Expanded clay [1] • Drying in an oven; 0.46
Ceramsite [20] • Coating with protection materials. 0.51

Scrap rubber tires are a major form of environmental solid waste that are dangerous not only due
to being a potential environmental threat, but also because they constitute a fire hazard and provide a
breeding ground for rats, mice, and mosquitoes [21,22]. One possible use for scrap tires is to crush
them into rubber particles and then incorporate them into concrete to decrease the brittle behavior
of the concrete. This results in a lower crack width in rubber concrete than in plain concrete [22,23].
Moreover, the rubber particles have the characteristics of surface roughness and high porosity, which
make them an attractive bacteria-carrier candidate. In this study, we investigate the feasibility of using
rubber particles to immobilize bacterial spores in self-healing concrete and to improve the anti-crack
provide a breeding ground for rats, mice, and mosquitoes [21,22]. One possible use for scrap tires is
to crush them into rubber particles and then incorporate them into concrete to decrease the brittle
behavior of the concrete. This results in a lower crack width in rubber concrete than in plain concrete
[22,23]. Moreover, the rubber particles have the characteristics of surface roughness and high porosity,
which make
Materials them
2019, 12, 2313 an attractive bacteria-carrier candidate. In this study, we investigate the feasibility
3 of 16
of using rubber particles to immobilize bacterial spores in self-healing concrete and to improve the
anti-crack capability of self-repairing concrete. Two types of self-healing concrete specimens with
capability of self-repairing
different rubber concrete.
particle sizes Two types
are prepared of self-healing
to quantify concrete effect.
the crack-healing specimens with different
The slump and the
rubber particle sizes are prepared to quantify the crack-healing effect. The slump and
mechanical properties of the self-healing rubber concrete are compared to those of plain concrete the mechanical
and
properties of the self-healing rubber concrete are compared to those of plain concrete
normal rubber concrete. In addition, a comparison of rubber particles to other bacteria carriers is and normal
rubber
performed.concrete. In addition, a comparison of rubber particles to other bacteria carriers is performed.

2. Materials and Methods

2. Materials and Methods
2.1. Materials
2.1. Materials
A ureolytic-type bacterium, Sporosarcina pasteurii ATCC11859, was used in this study. It was
A ureolytic-type bacterium, Sporosarcina pasteurii ATCC11859, was used in this study. It was
cultured in a sterilized liquid medium containing 20 g/L of yeast extract, 10 g/L of ammonium sulfate,
cultured in a sterilized liquid medium containing 20 g/L of yeast extract, 10 g/L of ammonium sulfate,
and 17.5 g/L of Tris regent. After cultivation in a shaker for 24 h at a temperature of 30 ◦ C and a
and 17.5 g/L of Tris regent. After cultivation in a shaker for 24 h at a temperature of 30 °C and a
rotational speed of 220 rpm, the bacteria reached a stable growth period. Then, the bacterial suspension
rotational speed of 220 rpm, the bacteria reached a stable growth period. Then, the bacterial
was centrifuged at 4000 rpm for 30 min to separate the bacterial cells from the cultivation solution.
suspension was centrifuged at 4000 rpm for 30 min to separate the bacterial cells from the cultivation
The supernatant was dispensed with and the remaining bacterial cells were re-suspended in 9 g/L of
solution. The supernatant was dispensed with and the remaining bacterial cells were re-suspended
NaCl solution. The re-suspended bacterial solution was placed in an oven at 30 ◦ C for 14 days, during
in 9 g/L of NaCl solution. The re-suspended bacterial solution was placed in an oven at 30 °C for 14
which time the bacteria remained in starvation. After that, almost all bacterial cells turned into bacterial
days, during which time the bacteria remained in starvation. After that, almost all bacterial cells
spores. This was confirmed by microscopic observation of the re-suspension solution as shown in
turned into bacterial spores. This was confirmed by microscopic observation of the re-suspension
Figure 1, in which the bacterial spores are dyed red. The concentration of spores in the bacterial spore
solution as shown in Figure 1, in which the bacterial spores are dyed red. The concentration of spores
suspension was determined to be 2.5E9 spores/mL by a plate count method [24]. The bacterial spore
in the bacterial spore suspension was determined to be 2.5E9 spores/mL by a plate count method [24].
suspension was stored at 4 ◦ C prior to the tests.
The bacterial spore suspension was stored at 4 °C prior to the tests.

Figure 1. A micrograph of bacterial spores.

Figure 1. A micrograph of bacterial spores.

Concrete specimens were prepared with ordinary Portland cement (P.O. 42.5, Tianjin, China),
Concrete specimens were prepared with ordinary Portland cement (P.O. 42.5, Tianjin, China),
tap water, local natural sand (distributed within 0~5 mm), and coarse aggregate (gravel, distributed
tap water, local natural sand (distributed within 0~5 mm), and coarse aggregate (gravel, distributed
within 5~25 mm). We adopted two rubber particle sizes (1~3 mm and 0.2~0.4 mm) in this study, whose
within 5~25 mm). We adopted two rubber particle sizes (1~3 mm and 0.2~0.4 mm) in this study,
particles size distributions were displayed in Figure 2. All of the rubber particles were obtained by
cutting scrap rubber tire casings, which contain 37% natural rubber, 25% butadiene rubber, 30% carbon
black, and 8% other additives. All other chemicals used in this study were of Analytical Reagent
(AR) grade.
Materials 2019, 12, x FOR PEER REVIEW 4 of 16

whose particles size distributions were displayed in Figure 2. All of the rubber particles were
obtained by cutting scrap rubber tire casings, which contain 37% natural rubber, 25% butadiene
Materials 2019, 12, 2313 4 of 16
rubber, 30% carbon black, and 8% other additives. All other chemicals used in this study were of
Analytical Reagent (AR) grade.

Figure 2.
Figure Particle size
2. Particle size distributions
distributions of
of the
the rubber
rubber particles.
particles.

2.2. Concrete Specimen Preparation

2.2. Concrete Specimen Preparation
Figure 3 shows the process for preparing the self-healing concrete specimens by immobilizing
Figure 3 shows the process for preparing the self-healing concrete specimens by immobilizing
bacterial spores in rubber particles. The details of the procedure are as follows:
bacterial spores in rubber particles. The details of the procedure are as follows:
1. The
1. Thebacterial
bacterialspore
sporesuspension
suspensionwas wasprepared
preparedaccording
accordingto tothe
the method
method described
described inin Section
Section 2.1.
2.1.
Then, the rubber particles were mixed with the bacterial spore
Then, the rubber particles were mixed with the bacterial spore suspension and 4% sodium suspension and 4% sodium
alginate(SA)
alginate (SA) solution.
solution. The
Theratio
ratioofofSA
SAsolution,
solution,bacterial
bacterialspore
spore suspension,
suspension, and and rubber
rubber particles
particles
was 1 mL:1 mL:2.2 g. The role of the SA solution was to adhere the bacterial
was 1 mL:1 mL:2.2 g. The role of the SA solution was to adhere the bacterial spores to the rubber spores to the
rubber particles and coat the rubber particles to protect the spores from leaving
particles and coat the rubber particles to protect the spores from leaving the pores in the rubber the pores in the
rubber particles.
particles. The proportion
The proportion of theseof these
threethree substances
substances resulted
resulted fromfrom ensuringthat
ensuring thatthere
there were
were
3.75E13 bacterial spores per m 3 of concrete and that there was sufficient SA solution to disperse
3.75E13 bacterial spores per m3 of concrete and that there was sufficient SA solution to disperse
thespores
the spores under
under the
the premise
premise that
that the
the amount
amount of of rubber
rubber particles
particles could
could be
be determined.
determined.
2. The
2. The mixture
mixture of
of the
thethree
threesubstances
substanceswas wasstirred
stirredand then
and dried
then in ain
dried cool and and
a cool ventilated placeplace
ventilated until
the SA had completely solidified.
until the SA had completely solidified.
3. Before preparing
3. Before preparing thetheself-healing
self-healingconcrete,
concrete, thethespore-containing
spore-containing rubber particles
rubber werewere
particles coated with
coated
cement
with (1 kg(1ofkg
cement rubber particles
of rubber withwith
particles 2 kg2ofkgcement, notenote
of cement, thatthat
this this
cement waswas
cement counted
countedin the
in
mixing proportion of the concrete) by a stirring method. Thus, the cement
the mixing proportion of the concrete) by a stirring method. Thus, the cement can protect SA can protect SA from
directly
from contacting
directly with free
contacting withwater
free and
waterslow anddown
slowthe dissolution
down of SA in the
the dissolution of fresh
SA inconcrete.
the fresh
4. concrete.
When manufacturing the self-healing concrete specimens, the nutrients for bacterial growth (yeast
4. Whenextractmanufacturing
and ammonium sulfate)
the and theconcrete
self-healing necessities for bacterial
specimens, the mineralization (calciumgrowth
nutrients for bacterial acetate
and urea)
(yeast were
extract mixed
and into the tap
ammonium water,
sulfate) andwhich was addedfor
the necessities to the concrete
bacterial mixture following
mineralization the
(calcium
coarse aggregate,
acetate and urea) cement,
were mixedrubber particles
into the tapand sand.which was added to the concrete mixture
water,
following the coarse aggregate, cement, rubber particles and sand.
Materials2019,
Materials 2019,12,
12,2313
x FOR PEER REVIEW 55ofof16
16

Figure 3. Preparation of self-healing concrete specimens via immobilization of bacterial spores in

rubber
Figureparticles.
3. Preparation of self-healing concrete specimens via immobilization of bacterial spores in
rubber particles.
Five types of concrete were prepared in this study: two types of self-healing rubber concrete, each
with aFive
different
types rubber particle
of concrete weresize (1~3 mm
prepared in(SRC-L) andtwo
this study: 0.2~0.4
typesmm of (SRC-S)),
self-healingplain concrete
rubber (C),
concrete,
and two types of normal rubber concrete (RC-L and RC-S). The latter three types
each with a different rubber particle size (1~3 mm (SRC-L) and 0.2~0.4 mm (SRC-S)), plain concreteof concrete were
used as contrasts
(C), and two types of of
thenormal
self-healing
rubberconcretes’ properties.
concrete (RC-L The water-to-cement
and RC-S). The latter threeratio
typesofofthese mixtures
concrete were
was 0.54. Note that the rubber particles replaced 10% of the volume of the sand in the normal
used as contrasts of the self-healing concretes’ properties. The water-to-cement ratio of these mixtures rubber
concretes
was 0.54. and
Notethe self-healing
that the rubberrubber
particlesconcretes
replacedto10%
effectively decrease
of the volume ofthe
the brittleness
sand in theofnormal
the concrete
rubber
and not reduce
concretes the self-healing
and the strength of the concrete
rubber too much
concretes at the same
to effectively time [25].
decrease theThe mixing proportions
brittleness of
of the concrete
the specimens are shown in Table 2. All specimens were cured under the conditions of 20 ± 2 ◦ C and
and not reduce the strength of the concrete too much at the same time [25]. The mixing proportions
higher
of the than 95% humidity
specimens are shown forin
different
Table 2.time
All periods.
specimens were cured under the conditions of 20 ± 2 °C
and higher than 95% humidity for different time periods.
Table 2. Mixing proportions of the concrete specimens.
Table 2. Mixing proportions of the concrete specimens.
Concrete Type C RC-L RC-S SRC-L SRC-S
CementConcrete
(kg/m3 ) Type 350 C 350 RC-L RC-S
350 SRC-L
350 SRC-S350
Sand 1 Cement
(kg/m3 ) (kg/m3) 750 350675 350 675350 675
350 350 675
3) 1110
Coarse aggregate
Sand(kg/m
3
1 (kg/m 3) 7501110 675 1110
675 1110
675 675 1110
Water (kg/m ) 190 190 190 190 190
Coarse aggregate (kg/m3) 1110 1110 1110 1110 1110
Yeast extract (kg/m3 ) 1.4 1.4
Water(kg/m
Ammonium sulfate (kg/m3 ) 3) 190 190190 190
0.7 190 0.7
Calcium Yeast
acetateextract
(kg/m3 (kg/m
3) 1.4 1.4 7
) 7
Ammonium 3
Urea (kg/m ) sulfate (kg/m3) 0.77 0.7 7
particles 2acetate
Rubber Calcium (kg/m3 )(kg/m3) 33 33 33
7 7 33
3 3.75 7× 1013 13
Bacterial spores (pores/m
Urea (kg/m) 3) 7 × 10
3.75
1 3 2 kg/m3 .
Rubber apparent density
particles of 2547
2 (kg/m 3) kg/m ; apparent
33 density
33 of 1120 33 33
Bacterial spores (pores/m ) 3 3.75E13 3.75E13
2.3. Bacterial Growth Test with SA and Rubber Particles
1 apparent density of 2547 kg/m3; 2 apparent density of 1120 kg/m3.
To determine the effects of rubber particles and SA on bacterial spore growth, six bottles of
100 mL of fresh
2.3. Bacterial liquidTest
Growth medium were
with SA andprepared then separately inoculated with: 1 mL of bacterial spore
Rubber Particles
suspension (I); 1 mL of bacterial spore suspension and 1 mL of SA solution (II); 1 mL of bacterial
To determine the effects of rubber particles and SA on bacterial spore growth, six bottles of 100
spore suspension and 2.2 g of sterilized rubber particles (III); 1 mL of bacterial spore suspension and
mL of fresh liquid medium were prepared then separately inoculated with: 1 mL of bacterial spore
2.2 g of unsterilized rubber particles (IV); 2.2 g of sterilized spore-containing rubber particles (V); and
suspension (I); 1 mL of bacterial spore suspension and 1 mL of SA solution (II); 1 mL of bacterial
2.2 g of unsterilized spore-containing rubber particles (VI). All media were cultured in a shaker at a
spore suspension and 2.2 g of sterilized rubber particles (III); 1 mL of bacterial spore suspension and
temperature of 30 ◦ C and a rotational speed of 220 rpm for four days. The urease activities in the
2.2 g of unsterilized rubber particles (IV); 2.2 g of sterilized spore-containing rubber particles (V); and
media were detected using a conductivity method [26] to estimate the amount of bacterial growth.
2.2 g of unsterilized spore-containing rubber particles (VI). All media were cultured in a shaker at a
temperature of 30 °C and a rotational speed of 220 rpm for four days. The urease activities in the
media were detected using a conductivity method [26] to estimate the amount of bacterial growth.
Materials 2019, 12, 2313 6 of 16
Materials 2019, 12, x FOR PEER REVIEW 6 of 16

2.4.
2.4. Crack Manufacturing
Manufacturing Methods
Methods and
and Self-Healing
Self-Healing Incubation
Incubation Conditions
Conditions
Three
Three prismatic
prismaticspecimens
specimens(100 (100mmmm×× 100100 mmmm × × 300 mm) were prepared to observe the the capacity
capacity
of
of each
each type
type of
of concrete
concrete to to self-heal
self-healcracks.
cracks. The
The cracks
cracks were
were created
created at
at the
the concrete
concreteage ageof
of 77 days
days using
using
aa three-point
three-pointbending
bendingmethod
method (see Figure
(see Figure4a).4a).
TheThespecimens
specimenswere loaded
were at a rate
loaded at a of 0.5ofmm/min
rate until
0.5 mm/min
auntil
cleara crack appeared
clear crack at theat
appeared subface. The three
the subface. specimens
The three of each
specimens type type
of each of concrete werewere
of concrete loaded to a
loaded
different extentextent
to a different to obtain crackscracks
to obtain with different widthswidths
with different distributed in the in
distributed range
the 0.2~0.9 mm. Four
range 0.2~0.9 mm.markFour
lines
markwere
linesdrawn on the on
were drawn cracked surfacesurface
the cracked of the concrete and the and
of the concrete intersections of the mark
the intersections lines
of the markandlines
the
cracks
and thewere used
cracks wereto monitor the progression
used to monitor of crack self-healing,
the progression as shownasinshown
of crack self-healing, Figurein 4a.Figure 4a.
After
Afterthethecracks
crackswerewerecreated,
created, thethe
concrete specimens
concrete werewere
specimens placed in a water
placed in a box for box
water crackforhealing
crack
at a temperature ◦
of 20 C,ofas20shown in Figure 4b. An4b.airAn
pump was used
healing at a temperature °C, as shown in Figure air pump was tousedprovide the oxygen
to provide that
the oxygen
the
thatconcrete needed
the concrete to self-heal.
needed The self-healing
to self-heal. incubation
The self-healing periodperiod
incubation lasted for 28 days,
lasted for 28and theand
days, results
the
of crackof
results repair
crackwere observed
repair at 3 days,
were observed at 73 days,
days, 14 days,14
7 days, and 28 days.
days, and 28 days.

(a)

(b)
Figure 4. The
Figure 4. The creation
creation of
of cracks
cracks (a)
(a) and
and the
the incubation
incubation conditions
conditions for
for the
the self-healing
self-healing concrete
concrete(b).
(b).

2.5. Concrete Properties Testing Methods

2.5. Concrete Properties Testing Methods
The slump of fresh concrete, the compressive strength of concrete, and the splitting tensile strength
The slump of fresh concrete, the compressive strength of concrete, and the splitting tensile
of concrete were investigated according to the Code for Hydraulic Concrete in China (SL 352-2006).
strength of concrete were investigated according to the Code for Hydraulic Concrete in China (SL
352-2006).
2.6. Compressive Strength Recovery Test
Three cubic specimens
2.6. Compressive (150 mm
Strength Recovery × 150 mm × 150 mm) of each type of concrete were pre-compressed
Test
to their maximum strengths at the age of 7 days. These specimens with many cracks were placed
underThree cubic specimens
the self-healing (150conditions
incubation mm × 150 formm × 150Afterward,
21 days. mm) of each
their type of concrete
compressive werewere
strengths pre-
compressed to their maximum strengths at the age of 7 days. These specimens with many cracks were
placed under the self-healing incubation conditions for 21 days. Afterward, their compressive
strengths were determined again. The strength regain of each type of concrete is expressed as the
average of three specimens, which was calculated by Equation (1).
Strength regain (%) = F/F0 × 100 (1)
Materials 2019, 12, 2313 7 of 16

determined again. The strength regain of each type of concrete is expressed as the average of three
specimens, which was calculated by Equation (1).

Strength regain (%) = F/F0 × 100 (1)

Materials 2019, 12, x FOR PEER REVIEW 7 of 16

where F and F0 are the compressive strength at the age of 28 days of the specimens with and without
where F and F0 are the compressive strength at the age of 28 days of the specimens with and without
precompression, respectively, in MPa.
precompression, respectively, in MPa.
3. Results
3. Results
3.1. Bacterial Growth under the Effect of SA and Rubber Particles
3.1. Bacterial Growth under the Effect of SA and Rubber Particles
Figure 5 displays the urease activities in the bacterial growth process under the effect of SA and
rubberFigure 5 displays
particles. the urease
The results showactivities in the bacterial
that the bacterial spores growth
germinatedprocess
and under the effect
proliferated ofsix
in all SAtests,
and
rubber particles.
producing ureaseThe results
in the show Test
medium. that (I)
thewas
bacterial sporestest,
the control germinated and proliferated
whose urease activity began in all
to six tests,
increase
producing urease in the medium. Test (I) was the control test, whose urease
rapidly (logarithmic phase) after a short adjustment phase, until a urease activity of about 12 mM activity began to increase
rapidly
urea/min(logarithmic
was reachedphase)
at 15 h.after
Then, a short adjustment
the urease activity phase, until a urease
was maintained activity
during of about
a stable phase12 mM
before
urea/min
it declinedwas reached
at about 70 at 15 h.decline
h (the Then, the urease
phase). activity
Tests was maintained
(II), (III), and (IV) were during
similara stable
to thephasecontrolbefore
test,
it declined at about 70 h (the decline phase). Tests (II), (III), and (IV) were
reaching the stable phase at about 15 h. However, the urease activities in the stable phase in Test (II)similar to the control test,
reaching
were aboutthe19
stable
mM phase
urea/min, at about 15 h. However,
58% higher than those theinurease activities
the control test,in theabout
and stable8 phase in Test (II)
mM urea/min in
were about 19 mM urea/min, 58% higher than those in the control test, and
Tests (III) and (IV), 33% lower than those in the control test. This indicates that SA was favorable for about 8 mM urea/min in
Tests (III) and
the urease (IV), 33%
activities and lower than particles
the rubber those in thewere control
adverse. test.In
This indicates
Tests (V) andthat (VI),SA
thewas favorable
urease for
activities
the
did urease activities
not increase untiland
15 the rubber
h after particles implying
inoculation, were adverse. In Tests
that the (V) andphase
adjustment (VI), the
wasurease
longeractivities
in these
did not increase until 15 h after inoculation, implying that the adjustment
two tests. This may be because it took a long time for the bacteria to adjust to the new substances phase was longer in these
two tests. This may be because it took a long time for the bacteria to adjust
in the environment. In addition, the bacterial spores in Tests (V) and (VI) were released slowly into to the new substances in
the
the environment.
medium because In addition,
of the coattheofbacterial sporesininaTests
SA, resulting lag in(V)theand (VI) were
increase releasedconcentration
of bacterial slowly into the in
medium because of the coat of SA, resulting in a lag in the increase of
the medium. At about 40 h, the urease activities in Tests (V) and (VI) reached the maximum value bacterial concentration in the
medium.
of about 12 AtmMabout 40 h, the
urea/min, urease
which was activities
similar to in that
Testsin(V) the and (VI)test,
control reached the maximum
suggesting value of
that the bacterial
about 12 mM urea/min,
spore immobilization whichis was
method similar
feasible and to
does thatnot inaffect
the control test, suggesting
the bacterial activities inthatthe the bacterial
stable phase.
spore immobilization method is feasible and does not affect the bacterial activities
Comparing Test (III) with Test (IV), as well as Test (V) with Test (VI), it can be found that the sterilization in the stable phase.
Comparing Test (III)
of rubber particles hadwith Test on
no effect (IV),
theas well asactivities.
bacterial Test (V) This withproves
Test (VI), it can be of
the feasibility found that the
this bacterial
sterilization of rubber particles had no effect on the
spore immobilization method for large-scale engineering applications. bacterial activities. This proves the feasibility of
this bacterial spore immobilization method for large-scale engineering applications.

Figure 5. Bacterial growth under the effect of sodium alginate (SA) and rubber particles.
Figure 5. Bacterial growth under the effect of sodium alginate (SA) and rubber particles.

3.2. Optical Determination of Crack-Healing Capacity

Figure 6 displays the microscopic images of representative cracks of the five types of specimens
at different healing times. Some white healing compound can be observed visibly in the cracks of all
of the specimens, implying that these five types of concrete all had some self-healing capacity.
However, this capacity was relatively limited in the C, RC-L, and RC-S specimens. After 28 days of
healing, the cracks in these three types of concrete were only partially healed, with an initial crack
Materials 2019, 12, 2313 8 of 16

3.2. Optical Determination of Crack-Healing Capacity

Figure 6 displays the microscopic images of representative cracks of the five types of specimens at
different healing times. Some white healing compound can be observed visibly in the cracks of all of the
specimens, implying that these five types of concrete all had some self-healing capacity. However, this
capacity was relatively limited in the C, RC-L, and RC-S specimens. After 28 days of healing, the cracks
in these three types of concrete were only partially healed, with an initial crack width of less than
0.45 mm. In the SRC-L and SRC-S specimens, the wider cracks (0.72 mm and 0.48 mm, respectively)
Materials 2019, 12, x FOR PEER REVIEW 8 of 16
were completely repaired.

Figure 6. The results of the optical determination of the crack-healing capacity.

Figure 6. The results of the optical determination of the crack-healing capacity.
Figure 7 illustrates the maximum values of the widths of completely healed cracks for the five
types of concrete
Figure at different
7 illustrates healing values
the maximum times. ofAs thethe healing
widths time increased,
of completely healedthe specimens
cracks for the with
five
no self-healing
types of concreteagent (C, RC-L,
at different andtimes.
healing RC-S)As exhibited improving
the healing values the
time increased, of maximum
specimens healed
with nocrack
self-
width, but
healing agentonly
(C,reached
RC-L, anda value
RC-S)ofexhibited
0.25 mm at 28 days. values
improving In contrast, the self-healing
of maximum rubberwidth,
healed crack concretes
but
(SRC-L
only and SRC-S)
reached a valueshowed
of 0.25 ammmuch better
at 28 days.healing capacity,
In contrast, the especially
self-healingtherubber
SRC-Lconcretes
specimens that had
(SRC-L anda
maximum
SRC-S) healed
showed crack better
a much width healing
of 0.86 mm at three
capacity, days. The
especially thereason
SRC-Lbehind this phenomenon
specimens was that
that had a maximum
the specimens
healed with of
crack width no 0.86
self-healing agent days.
mm at three were healed only by
The reason the hydration
behind of unhydrated
this phenomenon was thatcement
the
particles [7],
specimens butno
with theself-healing
self-healingagent
rubber concretes
were healedwere
onlyalso
by healed by the bacterial
the hydration mineralization.
of unhydrated cement
Therefore,
particles thebut
[7], self-healing technique
the self-healing using
rubber rubber particles
concretes were also to immobilize
healed by thebacterial spores
bacterial was effective
mineralization.
and greatlythe
Therefore, improved the self-healing
self-healing technique capacity of the concrete.
using rubber particles to immobilize bacterial spores was
To quantify
effective the crack-healing
and greatly improved the capacity, 10 measurement
self-healing capacity of the points were selected and averaged along
concrete.
with the length of each crack in the microscopic image. The crack widths at the measurement points
were measured. The crack healing percentage was determined as follows:

Healing percentage (%) = (d0 − dt)/d0 × 100 (2)

where d0 is the average initial crack width and dt is the average width measured at the healing time of
all 10 measurement points.
 SRC-S) showed a much better healing capacity, especially the SRC-L specimens that had a maximum
healed crack width of 0.86 mm at three days. The reason behind this phenomenon was that the
specimens with no self-healing agent were healed only by the hydration of unhydrated cement
particles [7], but the self-healing rubber concretes were also healed by the bacterial mineralization.
Therefore, the self-healing technique using rubber particles to immobilize bacterial spores was
Materials 2019, 12, 2313 9 of 16
effective and greatly improved the self-healing capacity of the concrete.

Materials 2019, 12, x FOR PEER REVIEW 9 of 16

To quantify the crack-healing capacity, 10 measurement points were selected and averaged
along with the length of each crack in the microscopic image. The crack widths at the measurement
points were measured. The crack healing percentage was determined as follows:
Healing percentage (%) = (d0 − dt)/d0 × 100 (2)
Figure 7. Maximum healed crack width at different healing times.
where d0 is the average initial crack width and dt is the average width measured at the healing time
ofFigure
all 10 measurement Figure 7. Maximum healed crack width at different healing times.
8 compares thepoints.crack healing percentage of the SRC-L concrete with that of the SRC-S concrete
Figure 8 compares the crack healing percentage of the SRC-L concrete with that of the SRC-S
at different healing times. At the healing time of three days (Figure 8a), half of the cracks in the SRC-L
concrete at different healing times. At the healing time of three days (Figure 8a), half of the cracks in
concrete specimens had been completely repaired from an initial width of 0.22~0.86 mm, and only three
the SRC-L concrete specimens had been completely repaired from an initial width of 0.22~0.86 mm,
cracks with an initial crack width of 0.22~0.54 mm in the SRC-S concrete had the healing percentage of 100%.
and only three cracks with an initial crack width of 0.22~0.54 mm in the SRC-S concrete had the
In addition, we found that the initial crack width had no influence on the healing percentage. For example,
healing percentage of 100%. In addition, we found that the initial crack width had no influence on
Crack A in the SRC-L concrete with a width of 0.86 mm was completely healed, but Crack B with a width of
the healing percentage. For example, Crack A in the SRC-L concrete with a width of 0.86 mm was
0.38completely
mm had a healing percentage
healed, but Crack Bof 11%.
with The widest
a width of 0.38crack
mm hadhad athe best healing
healing capacity,
percentage which
of 11%. Themay have
widest
resulted from the higher bacterial concentration in this crack. This proves that the
crack had the best healing capacity, which may have resulted from the higher bacterial concentrationdistribution of bacterial
carriers in crack.
in this the self-healing
This proves concrete
that thewas not uniform.
distribution As the time
of bacterial increased,
carriers the healing concrete
in the self-healing percentages wasof the
not
cracks
uniform. As the time increased, the healing percentages of the cracks continued to improve, as shownall
continued to improve, as shown in Figure 8b,c. When the healing time reached 28 days, almost
of the cracks 8b,c.
in Figure in theWhen
SRC-L theconcrete
healingweretime completely repaired
reached 28 days, andall
almost theofmaximum
the crackshealed crack width
in the SRC-L was
concrete
0.86were
mm.completely
The cracksrepaired
in the SRC-S
and the specimens
maximumallhealedreached healing
crack widthpercentages
was 0.86 mm. higher than 50%,
The cracks in theandSRC-the
maximum healed
S specimens all crack
reached width waspercentages
healing 0.52 mm. Ithigher
was clear
than that
50%,the
andcrack-healing
the maximumcapacity of thewidth
healed crack SRC-L
was 0.52
concrete wasmm. It was
better thanclear that
that of thethe crack-healing
SRC-S concrete, capacity
indicating ofthat
the SRC-L concrete
the rubber was with
particles bettera than
largerthat
size
wereof more
the SRC-S concrete,
suitable indicating that
for immobilizing the rubber
bacterial spores particles with a larger
in self-healing size were
concrete. This more
may be suitable
due tofor the
factimmobilizing bacterial
that larger rubber spores
particles caninprovide
self-healingmoreconcrete.
space forThis may be
bacterial due toavoiding
growth, the fact that larger
bacteria rubber
death due
particles can
to squeezing provide
in the more
concrete space for
matrix. bacterialthe
Moreover, growth, avoiding
bacteria can movebacteria
in thedeath due
cracks to squeezing
during in
the process
the concrete matrix. Moreover, the bacteria can move in the cracks during the
of incubation, providing calcium carbonate crystals to those locations that do not have carriers at the process of incubation,
providing
beginning. Thiscalcium
happened carbonate
in Crack crystals to those
C in Figure locations
8, which had that do not
a healing have carriers
percentage at in
of zero thethebeginning.
first seven
This happened in Crack C in Figure 8, which had a healing percentage of zero in
days after incubation, but was repaired gradually after the healing time of seven days. Bacterial transportthe first seven days
after incubation, but was repaired gradually after the healing time of seven days. Bacterial transport
in the cracks can help to heal cracks in locations with no bacterial carriers, improving the healing capacity
in the cracks can help to heal cracks in locations with no bacterial carriers, improving the healing
of the self-healing rubber concrete.
capacity of the self-healing rubber concrete.

(a) (b)

Figure 8. Cont.
 Materials 2019, 12, x FOR PEER REVIEW 10 of 16
Materials 2019,
Materials 12, 12,
2019, 2313x FOR PEER REVIEW 1010
of of
16 16

(c) (d)
(c) (d)
Figure 8. Crack healing percentage against the initial crack width at different healing times: three
Figure 8. Crack
Figure healing
8. Crack percentage
healing against
percentage thethe
against initial crack
initial width
crack at different
width healing
at different times:
healing three
times: days
three
days (a), seven days (b), 14 days (c), and 28 days (d).
(a),days
seven days
(a), (b),days
seven 14 days (c),
(b), 14 and(c),
days 28 and
days28(d).
days (d).

3.3.The
3.3. TheSlumps
Slumpsofofthe
theSelf-Healing
Self-HealingConcrete
Concrete
3.3. The Slumps of the Self-Healing Concrete
The
The The slumps
slumps
slumps ofof
ofthe
the the five
fivefive types
types of concrete
concrete
of concrete
types of are
areareshownshown
shown in ininFigure
Figure Figure It 9.
9. 9. can
It It
cancan
be be be
seen
seenseen
that
thatthat
the
the the
slumps
slumpsslumps
ofoftheof
the
RC-Lthe RC-L and RC-S concretes were 43 mm and 34 mm, respectively, which decreased by 6.5% andand
RC-L
and and
RC-S RC-S concretes
concretes were were
43 mm 43andmm 34 and
mm, 34 mm, respectively,
respectively, which which
decreased decreased
by 6.5% by
and 6.5%
26.1%,
26.1%,
26.1%,respectively,
respectively, comparedcompared
respectively, compared with
with the Cwith the CC(46
concrete
the concrete
mm). This
concrete (46mm).
(46 mm).
is due This is is
to the
This dueto to
ability
due ofthe
the the ability
rubber
ability ofparticles
of the the rubber
rubber to
particles
absorb the to
particles toabsorb
free waterthe
absorb the free
in the water
freefresh
water in the
concrete
in the fresh
fresh
into the concrete
poresinto
concrete into
on thepores
their
the pores
surface,onon their
reducing
their surface,
surface, reducing
the reducing
fluidity of
thethethe
fluidity
fresh of
mixture
fluidity the fresh
[21].
of the fresh mixture
Moreover, [21]. Moreover,
mixture the Moreover,
rubber particlesthe rubber
the rubberwith particles
a smallwith
particles with
sizeahada small
smalla size size
better had
hadcapacitya better capacity
to absorb
a better capacity
totoabsorb
water absorb water
than the water than
rubber the
the rubber
thanparticles
rubber particles
particles
with with
a largewithsize,aaaslarge
large
they size,
size,
had asas they
athey had
larger had a larger
a larger
surface surface
surface
area. Thearea. area.
SRC-S The
Theconcrete
SRC-SRC-
S Sconcrete
had aconcrete had
hadaa41.3%
41.3% higher 41.3% higher
slump higher
(65 mm) slump
slump
than (65(65mm)
the Cmm) thanthe
than
concrete, theCCconcrete,
whereasconcrete, whereas
whereas
the fresh mixturethetheof fresh
fresh mixture
themixture
SRC-L of the
of the
concrete
SRC-Lconcrete
SRC-L
collapsed concrete collapsed when
collapsed
when measuring measuring
measuring
the slump. These the
theslump.
slump.These
phenomena These phenomena
phenomena
indicate that the indicate
indicate
presence thatthat
thethe
of presence
the presence
healing
of
of the
agent the
led healing
healing agent
agent
to a more led to a more
led to mixture.
flowable flowable
flowable mixture.
mixture.
The previous The previous
The previous
studies studies
[17,27] show studies [17,27]
that[17,27] show
bacterial showthat bacterial
that bacterial
nutrients, such
nutrients,
asnutrients,
yeast suchurea,
such
extract, asyeast
as yeast extract,
calciumurea,
andextract, salt, and
urea, and
whichcalcium
were salt,
calcium salt,which
mixed withwere
which were
fresh mixed
mixed
concretewith infresh
with freshconcrete
concrete
self-healing in self-
in self-
concrete,
had healing
healing concrete,
concrete,
a negative effect had
hadonaacement
negative
negative effect
effect on
hydration onincement
cement hydration
the early age. ininthe
hydration theearly
earlyage.
age.

Figure9.9.The
Figure Theslumps
slumpsof
of the
the five
five types
types of
of concrete.
concrete.
Figure 9. The slumps of the five types of concrete.
3.4.3.4.
TheThe Compressive
Compressive Strengthofofthe
Strength theSelf-Healing
Self-HealingConcrete
Concrete
3.4. The Compressive Strength of the Self-Healing
typesConcrete
TheThe compressive
compressive strengthsofofthe
strengths thefive
fivetypes of concrete at
of at different
differentages
agesareareplotted
plottedininFigure
Figure10a.
10a.
WithWith
the the
The increase
compressive
increase of the
of the concrete’s
strengths
concrete’s of
age, age,
thethe the
five compressive
types strengths
of concrete
compressive of all
at different
strengths of all types
ages
types of concrete
ofare plotted
concrete improved
in Figureand
improved 10a.
the and
With the
ratethe rate of improvement
increase of theslowed
of improvement slowed
concrete’s
down down
age, gradually.
the compressive
gradually. The strengths
strengths
The strengths of the
of theofRC-L RC-L
all types and RC-S
of concrete
and RC-S concretes
improved
concretes at all
at were
and
ages all
theages
ratewere
lower lowerthose
than those
of improvement
than theofCthe
of slowed C concrete.
down
concrete. At At
28 28
gradually. days,
Thethe
days, thestrengths
strengths
strengths of RC-L
of the
of RC-L
RC-LandandRC-S
and RC-Sreached
RC-S concretes
reached
at25.1
25.1 all
MPa MPa
ages and and
were 26.8
26.8lowerMPa,
MPa, than respectively,
those of the
respectively, which were25.8%
C concrete.
which were 25.8% and 20.7%
At 28 and
days, lower
lowerthan
the strengths
20.7% the
thestrength
of RC-L
than and RC-S
strength ofof
the CC
reached
the
concrete (33.826.8
MPa), illustrating that the rubber particles had a20.7%
25.1
concreteMPa and
(33.8 MPa), MPa, respectively,
illustrating that thewhich
rubberwereparticles had a negative
25.8% and lower
negative effect
thanon
effect thestrength
the
on thecompressive
of the C
compressive
concreteof(33.8
strength MPa), illustrating
the concrete. This wasthat the rubber
because particles
the rubber had aas
particles, negative effect oncan
a soft material, the weaken
compressive
the
Materials2019,
Materials 12, x2313
2019, 12, FOR PEER REVIEW 11of
11 of16
16

strength of the concrete. This was because the rubber particles, as a soft material, can weaken the
compaction of the concrete. Moreover, the rubber particles with a large size, as larger weak points,
compaction of the concrete. Moreover, the rubber particles with a large size, as larger weak points,
led to a further reduction in strength compared to the rubber particles with a small size. At the age of
led to a further reduction in strength compared to the rubber particles with a small size. At the age
three days, the SRC-L concrete and the SRC-S concrete had compressive strengths that were lower
of three days, the SRC-L concrete and the SRC-S concrete had compressive strengths that were lower
than those of the C, RC-L, and RC-S concretes, proving that the degree of cement hydration in the
than those of the C, RC-L, and RC-S concretes, proving that the degree of cement hydration in the
self-healing concrete was low at early ages because of the negative effect of the nutrients on the cement
self-healing concrete was low at early ages because of the negative effect of the nutrients on the
hydration in the concrete. However, the compressive strengths of the SRC-L and SRC-S concretes
cement hydration in the concrete. However, the compressive strengths of the SRC-L and SRC-S
exceeded those of the corresponding normal rubber concrete by 16.4% and 0.7%, respectively, reaching
concretes exceeded those of the corresponding normal rubber concrete by 16.4% and 0.7%,
29.2 MPa and 27 MPa at the age of 28 days. In the self-healing rubber concretes, some bacterial spores
respectively, reaching 29.2 MPa and 27 MPa at the age of 28 days. In the self-healing rubber concretes,
were activated by the free water and produced calcites within the pores of the concrete matrix [28],
some bacterial spores were activated by the free water and produced calcites within the pores of the
improving the compaction of the concrete. In addition, the larger rubber particles provided a larger
concrete matrix [28], improving the compaction of the concrete. In addition, the larger rubber
space for bacterial mineralization, resulting in a greater compressive strength improvement in the
particles provided a larger space for bacterial mineralization, resulting in a greater compressive
SRC-L concrete. Nevertheless, the compressive strength of the SRC-L concrete at 28 days was still
strength improvement in the SRC-L concrete. Nevertheless, the compressive strength of the SRC-L
13.6% lower than that of the C concrete.
concrete at 28 days was still 13.6% lower than that of the C concrete.
Figure 10b shows the compressive strength regain of the five types of concrete. The C, RC-L,
Figure 10b shows the compressive strength regain of the five types of concrete. The C, RC-L, and
and RC-S concrete had a strength regain of 85%, 88%, and 86%, respectively, resulting from the
RC-S concrete had a strength regain of 85%, 88%, and 86%, respectively, resulting from the hydration
hydration of unhydrated cement particles in the concretes without self-healing agent. The strength
of unhydrated cement particles in the concretes without self-healing agent. The strength regains of
regains of the SRC-L and SRC-S concretes reached 96% and 90%, respectively, implying that calcites
the SRC-L and SRC-S concretes reached 96% and 90%, respectively, implying that calcites were
were produced in the cracks and improved the degree of repair in the self-healing concrete.
produced in the cracks and improved the degree of repair in the self-healing concrete.

(a) (b)

Figure10.
Figure Thecompressive
10.The compressivestrengths
strengths(a)
(a)and
andthe
thestrength
strengthregain
regain (b)
(b) of
of the
the five
five types
types of
of concrete.
concrete.

3.5. The
3.5. TheSplitting
SplittingTensile
Tensile Strength
Strength of
of the
the Self-Healing
Self-Healing Concrete
Concrete
The splitting
The splitting tensile
tensile strengths
strengthsof ofthe
thefive
fivetypes
typesofofconcrete
concrete atat
different
different ages areare
ages plotted in Figure
plotted 11a.
in Figure
It can be seen that the RC-L and RC-S concretes had splitting tensile strengths
11a. It can be seen that the RC-L and RC-S concretes had splitting tensile strengths that were lower that were lower than
that of
than theofCthe
that concrete during
C concrete the curing
during time,time,
the curing and were reduced
and were by 8.8%
reduced and 6.3%,
by 8.8% respectively,
and 6.3%, at the
respectively,
age of 28 days. The reason may be that the weak bonding between the rubber
at the age of 28 days. The reason may be that the weak bonding between the rubber particles and the particles and the cement
accelerated
cement the concrete’s
accelerated breakdown
the concrete’s [21]. The
breakdown twoThe
[21]. types
twooftypes
self-healing concreteconcrete
of self-healing had low hadsplitting
low
tensile strengths at three days because of the low cement hydration at early ages.
splitting tensile strengths at three days because of the low cement hydration at early ages. However, However, the splitting
tensile
the strengths
splitting of the
tensile self-healing
strengths of theconcrete began
self-healing to exceed
concrete those
began to of the corresponding
exceed normal rubber
those of the corresponding
concretes from the age of 14 days. At the age of 28 days, the splitting
normal rubber concretes from the age of 14 days. At the age of 28 days, the splitting tensile tensile strengths of the SRC-L
strengths
andthe
of SRC-S concretes
SRC-L were 2.55
and SRC-S MPa andwere
concretes 2.44 MPa,
2.55 respectively,
MPa and 2.44 whichMPa,is anrespectively,
improvementwhichof 17.5% is and
an
9.4%, respectively, over the corresponding normal rubber concretes (2.17
improvement of 17.5% and 9.4%, respectively, over the corresponding normal rubber concretes (2.17 MPa and 2.23 MPa), and an
improvement
MPa of 7.1%
and 2.23 MPa), and
and an2.5%, respectively,
improvement overand
of 7.1% the C concrete
2.5%, (2.38 MPa).
respectively, overThis
the Cmay be because
concrete (2.38
the SA coating on the rubber particles in the self-healing concrete can
MPa). This may be because the SA coating on the rubber particles in the self-healing concrete function as a coupling agent to
can
improve the bonding of rubber particles and cement.
function as a coupling agent to improve the bonding of rubber particles and cement.
Materials 2019, 12, 2313 12 of 16
Materials 2019, 12, x FOR PEER REVIEW 12 of 16

Figure
Figure 11b
11b shows
shows the
the crack
crack characteristics
characteristics of
of the
the C, RC-L,
RC-L, and SRC-L concretes
concretes after
after the
the splitting
splitting
tensile strength
tensile strength tests. The RC-L and SRC-L concretes had more narrow cracks than the C concrete,
proving
proving that
that the concretes with rubber particles had lower brittleness and
and better
better anti-crack
anti-crack capability.
capability.

(a)

(b)
Figure 11. The
Figure 11. The splitting
splittingtensile
tensilestrengths
strengthsofofthe
thefive
fivetypes
typesofofconcrete (a)(a)
concrete and thethe
and crack characteristics
crack of
characteristics
the concretes
of the (b).(b).
concretes

4. Discussion
4. Discussion
4.1. The Working Mechanism of the Self-Healing Rubber Concrete
4.1. The Working Mechanism of the Self-Healing Rubber Concrete
The test results have proved that the self-healing concrete using rubber particles to immobilize
Thespores
bacterial test results have proved
can produce that the in
precipitations self-healing
the cracks toconcrete using The
repair them. rubber particles
working to immobilize
mechanism of the
bacterial spores can produce precipitations in the cracks to repair them. The working mechanism
self-healing rubber concrete is displayed in Figure 12. When the concrete was not cracked, the bacterial of
the self-healing
spores rubber
were dormant in concrete
the poresisofdisplayed
the rubber inparticles,
Figure 12. When
which the coated
were concretebywas
SA. not cracked,
After the
the cracks
bacterial spores were dormant in the pores of the rubber particles, which were coated
appeared on the surface of the concrete, the oxygen and water in the external environment can enter by SA. After
the cracks
into appeared
the concrete on thethe
through surface of the
cracks. Theconcrete,
SA, coatedthe oxygen and water
on the rubber in theon
particles external environment
the surfaces of the
cracks, was dissolved slowly by the water and the bacterial spores were released fromthe
can enter into the concrete through the cracks. The SA, coated on the rubber particles on thesurfaces
rubber
of the cracks,
particles. Thesewas dissolved
bacterial slowly
spores wereby the water
activated andoxygen
by the the bacterial spores
and water were
in the released
cracks from the
and began to
rubber particles. These bacterial spores were activated by the oxygen and water in the
grow, using the nutrients in the concrete. Then, calcite precipitation was produced by the bacterial cellscracks and
began
in to grow,
the cracks andusing thethe
sealed nutrients in the concrete.
cracks gradually. In thisThen, calcite
process, some precipitation was
bacterial cells canproduced by the
be transported
bacterial cells in the cracks and sealed the cracks gradually. In this process, some bacterial cells can
with the water flow to locations with no bacterial carriers to produce crystals in the cracks.
be transported with the water flow to locations with no bacterial carriers to produce crystals in the
cracks.
Materials2019,
Materials 2019,12,
12,2313
x FOR PEER REVIEW 13of
13 of16
16

Figure 12. The working mechanism of the self-healing rubber concrete.

Figure 12. The working mechanism of the self-healing rubber concrete.
In the self-healing concrete, the role of the rubber particles is to:

• In the self-healing
protect the bacterialconcrete, the role
spores from of the rubber
squeezing particles
and high is to:
alkalinity in the concrete;
• provide spaces
protect the for bacterial
bacterial spores growth and mineralization;
from squeezing and
and high alkalinity in the concrete;
• reduce
providethe brittleness
spaces of the concrete
for bacterial and mineralization;
growth and improving the anti-crack
and capability of the concrete.
 reduce the brittleness of the concrete and improving the anti-crack capability of the concrete.
The role of the SA is to:
The role of the SA is to:
• wrap the bacterial spores in the pores of the rubber particles;
• wrap the
slowly bacterial
dissolve andspores
releaseinthe
thebacterial
pores ofspores
the rubber
whenparticles;
cracks appear; and
• improve the bonding of rubber particles and cement. cracks appear; and
slowly dissolve and release the bacterial spores when
 improve the bonding of rubber particles and cement.
4.2. Comparison of Rubber Particles with other Bacterial Carriers
4.2. Comparison of Rubber Particles with other Bacterial Carriers
A comparison between rubber particles and two other representative bacterial carriers is exhibited
A comparison
in Table 3. It is obvious between
that therubber
methodparticles
that usesand twoparticles
rubber other representative
to immobilize bacterial carriershas
bacterial spores is
exhibited in Table 3. It is obvious that the method that uses rubber particles to
the merits of a simple procedure, savings on bacterial spores, low cost, and better anti-crack capability immobilize bacterial
sporescompared
when has the merits of a simplemicrocapsules
with melamine procedure, savings on bacterial
and expanded spores,
perlite. The low cost, and
bacterial better
spore anti-
density
crack capability when compared with melamine microcapsules and expanded
in the concrete of this study is reduced by an order of magnitude in comparison to the other two perlite. The bacterial
spore density
bacterial in the
carriers; concretethe
however, of maximum
this study is reduced
width by cracks
of the an order thatof can
magnitude
be repairedin comparison
is similar to to the
the
other two bacterial carriers; however, the maximum width of the cracks that
others. This can bring about great savings on bacterial spores in practical engineering applications.can be repaired is similar
to addition,
In the others.the This
cost ofcanthebring
rubberabout great
particles savings
is very cheaponandbacterial
the price spores
of thein practical
bacterial engineering
carrier (rubber
applications. In addition,3 the cost of the rubber particles is very cheap and the
particles and SA) per m concrete is only 8.26 US$ (based on the price in Tianjin, China), which is lower price of the bacterial
carrier
than the(rubber
price ofparticles
the other and SA)bacterial
two per m3 concrete
carriers.isMoreover,
only 8.26 US$ (based on the
the anti-crack price inof
capability Tianjin, China),
the concrete
which
can is lower than
be improved therubber
by the price of the other
particle two bacterial
method, as showncarriers.
in Figure Moreover, the anti-crack
11b. Therefore, the use capability
of rubber
of the concrete can be improved by the rubber particle method, as shown
particles to immobilize bacterial spores in self-healing concrete has the potential to be more widelyin Figure 11b. Therefore,
the use
used forof rubber engineering
practical particles to immobilize
applications. bacterial
However, spores in self-healing
it was concrete
difficult to obtain has thedistribution
a uniform potential to
berubber
of more particles
widely usedin thefor practical
concrete engineering
because applications.
their density However,
is much lower it was
than that difficult
of sand. Thistolimitation
obtain a
uniform distribution of rubber particles
needs to be further considered in future research. in the concrete because their density is much lower than that
of sand. This limitation needs to be further considered in future research.
Materials 2019, 12, 2313 14 of 16

Table 3. Comparison of the rubber particles to other bacterial carriers.

Bacteria Carrier Melamine Microcapsule [16]
• A patented
polycondensation
reaction-based
microencapsulation process;
Immobilization procedure
• Melamine-based,
and contains an inert
substance to protect
the spores;
Expanded Perlite [7] Rubber Particle
• Impregnating the
expanded perlite
particles under
vacuum with the
bacterial
• Mixing the rubber
spore suspension;
particles, the SA solution,
• Drying the particles
and bacterial spore
in an oven;
suspension together;
• Spraying the
• Stirring them adequately;
nutrients onto the
• Drying in a cool and
surface of
ventilated place.
the particles;
• Coating the
particles with a
geopolymer coating

Bacterial spore density 1.8 × 1014 cells/m3 concrete 5 × 1014 cells/m3 concrete 3.75 × 1013 cells/m3 concrete
Maximum crack width 0.97 mm 0.79 mm 0.86 mm
Price Expensive 9.18 US$/m3 concrete 8.26 US$/m3 concrete
Other characteristics Higher anti-crack capability

5. Conclusions
This paper investigated the feasibility of using rubber particles as a novel bacterial carrier in
self-healing concrete. Two types of self-healing concrete were prepared with different rubber particle
sizes, whose healing capacities were evaluated. The properties of the self-healing rubber concrete,
including the slump, the compressive strength, and the splitting tensile strength, were compared with
those of plain concrete and normal rubber concrete. The following conclusions can be drawn:

• The method for immobilizing bacterial spores with rubber particles was feasible and the maximum
values of the widths of completely healed cracks after 28 days of healing were 0.86 mm in the
SRC-L concrete and 0.52 mm in the SRC-S concrete.
• The healing capacity of the SRC-L concrete was better than that of the SRC-S concrete because
larger rubber particles can provide larger spaces for bacteria to grow and mineralize, avoiding
bacteria death due to squeezing in the concrete.
• The compressive strength of the self-healing concrete was low at early ages due to the negative
effect of the nutrients in the concrete on cement hydration, but exceeded the compressive strength
of the corresponding normal rubber concrete at 28 days because calcites were produced within
the pores of the concrete matrix.
• The splitting tensile strength of the self-healing concrete was higher than that of the plain concrete,
resulting from SA improving the bonding between rubber particles and cement. In addition,
the rubber particles can enhance the anti-crack capability of the concrete.
• The immobilization of bacterial spores using rubber particles has the advantages of a simple
procedure, materials savings, and low cost, and provides a more suitable method for engineering
applications of self-healing concrete.

Even through there are some problems with self-healing rubber concrete that need to be solved,
such as the distribution of bacterial carriers in the concrete, the most suitable rubber particle size,
and the protective effect of rubber particles on bacterial spores in the long term, the use of rubber
particles remains an innovative and promising method for the immobilization of bacterial spores in
self-healing concrete.

Author Contributions: Conceptualization, H.X., J.L., and Y.Y.; methodology, H.X. and Y.Y.; validation, Y.Y.;
formal analysis, M.G.; investigation, H.X. and M.G.; resources, J.L.; data curation, M.G.; writing—original draft
preparation, H.X.; writing—review and editing, D.F. and Y.Y.; visualization, H.X.; project administration, Y.Y.;
funding acquisition, J.L.
Materials 2019, 12, 2313 15 of 16

Funding: This research was funded by the National Key R&D Program of China (Grant NO. 2016YFC0401905),
the National Natural Science Foundation of China (Grant NO. 51709198), and the Tianjin Natural Science
Foundation of China (Grant NO. 16JCQNJC07900).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Wiktor, V.; Jonkers, H.M. Quantification of crack-healing in novel bacteria-based self-healing concrete.
Cem. Concr. Compos. 2011, 33, 763–770. [CrossRef]
2. Luo, M.; Qian, C.X.; Li, R.Y. Factors affecting crack repairing capacity of bacteria-based self-healing concrete.
Constr. Build. Mater. 2015, 87, 1–7. [CrossRef]
3. Palin, D. A bacteria-based bead for possible self-healing marine concrete applications. Smart Mater. Struct.
2016, 25, 084008. [CrossRef]
4. Jing, X.; Wu, Y. Multiscale mechanical quantification of self-healing concrete incorporating non-ureolytic
bacteria-based healing agent. Cem. Concr. Res. 2014, 64, 1–10.
5. Muhammad, N.Z.; Shafaghat, A.; Keyvanfar, A.; Majid, M.Z.A.; Ghoshal, S.K.; Yasouj, S.E.M.; Ganiyu, A.A.;
Kouchaksaraei, M.S.; Kamyab, H.; Taheri, M.M. Tests and methods of evaluating the self-healing efficiency
of concrete: A review. Constr. Build. Mater. 2016, 112, 1123–1132. [CrossRef]
6. Wang, J.; Tittelboom, K.V.; Belie, N.D.; Verstraete, W. Use of silica gel or polyurethane immobilized bacteria
for self-healing concrete. Constr. Build. Mater. 2012, 26, 532–540. [CrossRef]
7. Zhang, J.; Liu, Y.; Feng, T.; Zhou, M.; Zhao, L.; Zhou, A.; Li, Z. Immobilizing bacteria in expanded perlite for
the crack self-healing in concrete. Constr. Build. Mater. 2017, 148, 610–617. [CrossRef]
8. Huang, H.; Ye, G. A Review on Self-Healing in Reinforced Concrete Structures in View of Serving Conditions.
In Proceedings of the 3rd International Conference on Service Life Design for Infrastructure, Zhuhai, China,
15–17 October 2014.
9. Huang, H.; Ye, G.; Qian, C.; Schlangen, E. Self-healing in cementitious materials: Materials, methods and
service conditions. Mater. Des. 2016, 92, 499–511. [CrossRef]
10. Farrugia, C.R.; Borg, R.P.; Ferrara, L.; Buhagiar, J. The Application of Lysinibacillus sphaericus for surface
treatment and crack healing in mortar. Front. Built Environ. 2019, 5, 1–10. [CrossRef]
11. Azarsa, P.; Gupta, R.; Biparva, A. Assessment of self-healing and durability parameters of concretes
incorporating crystalline admixtures and Portland Limestone Cement. Cem. Concr. Compos. 2019, 99, 17–31.
[CrossRef]
12. Alazhari, M.; Sharma, T.; Heath, A.; Cooper, R.; Paine, K. Application of expanded perlite encapsulated
bacteria and growth media for self-healing concrete. Constr. Build. Mater. 2018, 160, 610–619. [CrossRef]
13. Jing, X.; Wang, X. Self-healing of concrete cracks by use of bacteria-containing low alkali cementitious
material. Constr. Build. Mater. 2018, 167, 1–14.
14. Chen, X.; Yuan, J.; Alazhari, M. Effect of microbiological growth components for bacteria-based self-healing
on the properties of cement mortar. Materials 2019, 12, 1303. [CrossRef] [PubMed]
15. De Araujo, M.A.P.G.; Chatrabhuti, S.; Gurdebeke, S.; Alderete, N.M.; Van Tittelboom, K.; Raquez, J.;
Cnudde, V.; Van Vlierberghe, S.; De Belie, N.; Gruyaert, E. Poly(methyl methacrylate) capsules as an
alternative to the ‘’proof-of-concept” glass capsules used in self-healing concrete. Cem. Concr. Compos. 2018,
89, 260–271. [CrossRef]
16. Wang, J.; Soens, H.; Verstraete, W.; De Belie, N. Self-healing concrete by use of microencapsulated bacterial
spores. Cem. Concr. Res. 2014, 56, 139–152. [CrossRef]
17. Dong, B.; Fang, G.; Ding, W.; Liu, Y.; Zhang, J.; Han, N.; Feng, X. Self-healing features in cementitious
material with urea–formaldehyde/epoxy microcapsules. Constr. Build. Mater. 2016, 106, 608–617. [CrossRef]
18. Wang, J.; Snoeck, D.; Van Vlierberghe, S.; Verstraete, W.; De Belie, N. Application of hydrogel encapsulated
carbonate precipitating bacteria for approaching a realistic self-healing in concrete. Constr. Build. Mater.
2014, 68, 110–119. [CrossRef]
19. Wang, J.Y.; Belie, N.D.; Verstraete, W. Diatomaceous earth as a protective vehicle for bacteria applied for
self-healing concrete. J. Ind. Microbio. Biotechnol. 2012, 39, 567–577. [CrossRef] [PubMed]
20. Xu, J.; Wang, B. Research on self-healing of concrete cracks by ceramsite immobilized microorganism.
Mater. Rev. 2017, 31, 127–131.
Materials 2019, 12, 2313 16 of 16

21. Siddique, R.; Naik, T.R. Properties of concrete containing scrap-tire rubber—an overview. Waste Manag. 2004,
24, 563–569. [CrossRef] [PubMed]
22. Thomas, B.S.; Gupta, R.C. A comprehensive review on the applications of waste tire rubber in cement
concrete. Renew. Sustain. Energy Rev. 2016, 54, 1323–1333. [CrossRef]
23. Khaloo, A.R.; Dehestani, M.; Rahmatabadi, P. Mechanical properties of concrete containing a high volume of
tire–rubber particles. Waste Manag. 2008, 28, 2472–2482. [CrossRef] [PubMed]
24. Allievi, L.; Möller, F. A method based on plate count for enumerating N2O/N2-producing bacteria from
nitrate in the soil. J. Basic Microbiol. 2010, 32, 291–298. [CrossRef]
25. Feng, L.; Yuan, Q.; Ma, Y.; Shi, C. Experimental study on mechanical property of rubber concrete.
J. Yangtze River Sci. Res. Instit. 2015, 32, 115–118.
26. Whiffin, V.S. Microbial CaCO3 Precipitation for the Production of Biocement. Ph.D Thesis, Murdoch
University, Perth, WA, Australia, 2004.
27. Tziviloglou, E.; Wiktor, V.; Jonkers, H.M.; Schlangen, E. Bacteria-based self-healing concrete to increase liquid
tightness of cracks. Constr. Build. Mater. 2016, 122, 118–125. [CrossRef]
28. Siddique, R.; Singh, K.; Kunal; Singh, M.; Corinaldesi, V.; Rajor, A. Properties of bacterial rice husk ash
concrete. Constr. Build. Mater. 2016, 121, 112–119. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).