Construction and Building Materials 232 (2020) 117222

Contents lists available at ScienceDirect

Construction and Building Materials

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

Sugar-coated expanded perlite as a bacterial carrier for crack-healing

concrete applications
Lu Jiang a,c, Guanhua Jia b,c, Chen Jiang a, Zhu Li a,c,⇑
a
College of Civil Engineering, Taiyuan University of Technology, Taiyuan 030024, PR China
b
Department of Civil Engineering, Shanxi University, Tauyuan 030013, PR China
c
Shanxi Shengke Microorganism Building Materials Technology Co., Ltd, Taiyuan 030032, PR China

h i g h l i g h t s

Adopting the sugar-coating method to immobilize bacteria and nutrients.

EP particles after wrapping can enhance the crack-healing capacity of concrete.

The crack-healing of concrete peaked when EP particles wrapped with MKPC.

The values of completely healed crack widths were up to 1.24 mm.

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

Article history: Microbial-induced carbonate precipitation (MICP) can realize self-diagnosis and self-healing of concrete
Received 7 June 2019
1-2 to reduce the maintenance costs and extend the lifetime of concrete structures. However, the self-healing
Received in revised form 19 September capacity of the cracks is below par using existing self-healing methods. Thus, adopting the sugar-coating
2 notes:
2019
method to immobilize bacteria and nutrients was proposed in this paper. The effect of expanded perlite
Accepted 11 October 2019
(EP) wrapped with various materials as bacterial carrier and nutrient carrier on the crack-healing capac-
3-4 ity of concrete has been investigated. The results indicated that after EP particles were coated, the numer-
2 notes: Keywords: ical tube pressure was increased, and the water absorption was reduced. Wrapping materials can
Microbial-induced carbonate precipitation
effectively reduce the breakage rate of the EP particles and resist the invasion of water in the process
Concrete of mixing with concrete. The EP particles after wrapping process were found to enhance the healing
5-6 Self-healing capacity of concrete cracks. In particular, the crack-healing of concrete peaked when EP particles immo-
2 notes: Bacteria bilized with bacterial spores were wrapped with a low-alkaline material (potassium magnesium phos-
Crack-healing phate cement). This result was also validated by the water permeability experiments. After 28 d of
Sugar-coating crack-healing, the maximum value of the completely healed crack width was 1.24 mm. EP is a low-
07 cost material; thus, it can be used as a bacterial carrier for large-scale self-healing concrete applications.
Ó 2019 Published by Elsevier Ltd.
s Agudelo Gallo

1. Introduction repaired before they develop into macrocracks, the safety of struc-
tures will be greatly enhanced, and the service life will be
Concrete is the most widely used building material due to its extended. Traditional crack-repairing method and passive repair
desirable properties such as high compressive strength, good dura- based on microbially mineralized deposition involve intensive
bility, and low cost. However, as a multiphase heterogeneous labor, making them inapplicable in specific concrete structures
material, concrete has relatively low tensile strength, high brittle- [3,4].
ness, and poor cracking resistance, making its performance easily Currently, commonly used repairing methods in engineering
affected by the surrounding complex environment in the long construction include surface repair, pressure grouting, filling, and
term; thus, it is difficult to avoid small cracks on the surface and concrete replacement. However, these passive restoration tech-
inside of concrete [1,2]. If these microcracks can be effectively nologies can no longer meet the requirements of modern buildings
for concrete materials. Therefore, the importance of developing
novel ways to repair cracks in concrete is paramount [5–7].
⇑ Corresponding author at: College of Civil Engineering, Taiyuan University of
Recently, with the development of concrete technology and
Technology, Taiyuan 030024, China.
E-mail address: [email protected] (Z. Li).
microbiological technology, the microbial-induced mineralization

https://doi.org/10.1016/j.conbuildmat.2019.117222
0950-0618/Ó 2019 Published by Elsevier Ltd.
2 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222

technique and its application to repair concrete cracks have driven

great attention. Compared with other repairing methods, it has the
following outstanding advantages [4,5,8–11]: (1) Microorganisms
use Bacillus alkalophilus to survive in the form of spores in the
uncracked concrete interior. Studies have shown that bacterial
spores can survive for more than 200 years in dry environments;
thus, they can meet the service life requirements of concrete struc-
tures. (2) Bacteria are evenly distributed within the concrete struc-
ture. Once cracks occur in concrete, water (or water and oxygen) Fig. 1. Self-healing agent.
enters into the concrete, microorganisms wake up, and microbial
metabolic mineralization deposits fill the cracks; thus, self-
ensuring sufficient mechanical properties and durability of con-
diagnosis and self-repair of concrete are realized. (3) Microbial-
crete. Here we demonstrate the sugar-coating approach to immo-
mineralization reaction produces calcium carbonate, which is not
bilize bacteria and nutrients. First, the bacteria were immobilized
only an environmentally friendly and durable material, but it also
in the carrier. Then, the nutrients were evenly wrapped on the
provides excellent compatibility with concrete materials.
outer surface of the carrier. Finally, the carrier was coated with a
Currently, studies concerning self-healing methods based on
protective shell, as shown in Fig. 1. Expanded perlite was used as
microbial mineralization mainly focus on bacterial and nutrient
a carrier for bacteria. The effects of different coating materials on
selection, bacterial immobilization methods, and investigate self-
the physical and mechanical properties of expanded perlite and
healing effects on the durability and mechanical properties of con-
the effects of the carrier package on the crack-healing of concrete
crete. Studies have shown that after a period of repair and mainte-
have been investigated.
nance, concrete crack with a width less than 0.3 mm could be
completely repaired by the microbial mineralization-based self-
repairing method; while for the increasing crack width, the self- 2. Materials and methods
healing method falls short of completing crack-healing [4,11–13].
In addition, for most existing microbial mineralization methods, 2.1. Expanded perlite
microorganisms are only deposited on the surface of cracks [14].
This phenomenon indicates that the microbial mineralization- Expanded perlite is abundant in nature, and China is at the fore-
induced self-healing method proposed by researchers has excellent front of the expanded perlite reserves in the world. EP is also
microcrack repair capability; however, repairing cracks with width attractive due to its low cost and convenient processing. Expanded
size more than 0.8 mm via this technique remains challenging. perlite is widely used as amorphous, lightweight mineral aggre-
Researchers used inorganic porous materials such as expanded gate, which is made of acidic volcanic glass lava inflated by instant
clay particles, diatomite, ceramsite, zeolite, and expanded perlite high temperature (1000–1300 °C), resulting in a large number of
as bacterial carriers to protect bacteria from the highly alkaline bubbles in its internal crystalline water. The EP particles were
environment of concrete [4,5,15,16]. These inorganic porous mate- obtained from Henan province and had a water absorption capac-
rials have rich interconnected pore-like structures, which can cre- ity of 366% under vacuum, a loose dry bulk density of 78.9 kg/m3,
ate the required environment for bacterial growth and metabolism. and particle size distribution 2–4 mm. Under atmospheric pres-
The relatively dense structure of the surface also separates the bac- sure, the water absorption of EP particles is 275% (in 24 h), and
teria from the highly alkaline environment of the concrete. the numerical tube pressure is 152.5 kPa. The cost of EP particles
Although these inorganic porous materials and microcapsules pro- is less than US$ 0.22 per kg in China [18].
vide effective protection for microbial growth and metabolism,
they usually have low numerical tube pressure. 2.2. Self-healing agent preparation
Nutrients are important components to ensure that the micro-
bial mineralization process can be carried out efficiently. It pro- An alkalophilic bacterium, Bacillus cohnii (DSM 6307), was pur-
vides a long-lasting supply of organic matter and calcium source chased from Be Na Culture Collection, China. B. cohnii was used due
to microbial mineralization. When nutrients are directly added to to its alkalophilic activity and lack of pathogenicity. The culture
concrete, the hydration reaction of cement may be significantly medium for spore production of B. cohnii included sucrose (3.5 g),
affected by nutrients, and may adversely affect the mechanical yeast extract (4 g), KH2PO4 (0.02 g), CaCl2 (0.166 g), KCl (0.476 g),
properties and durability of concrete [4,5,7,15,17–19]. To avoid MgSO47H2O (0.2 g), MnSO47H2O (0.02 g), Na2CO3 (4.2 g), and
the adverse effects of directly incorporating nutrients to cement NaHCO3 (5.3 g), each per liter of distilled water. The germination
hydration, some researchers immobilized bacteria and nutrients rates of spores at 90 min were 31.2, 73.6, 89.4, 43.3, and 10.5% at
into the same group of carriers (i.e., bacterial carriers) [4,20–22]. pH 7, 8, 9, 10, and 11, respectively. Mineralization activity of ger-
When the nutrients and bacteria used a shared carrier, the bacteria minated bacterial spores was achieved by detecting the concentra-
inevitably came into contact with the nutrients during the immo- tion of calcium ions using EDTA titrimetric method [24]. When the
bilization process, which may cause the microbial spores to pro- isovolumetric bacterial solution was added to the standard solu-
duce a small amount of mineralization reaction. Moreover, due to tion of calcium ions (final concentration of calcium ions was about
the limited porosity of the carrier, it is difficult to increase the con- 10 mM), the concentration of calcium ions was decreased by 78.3%
centration of microorganisms and nutrients in the crack area with after 7 days of mineralization.
the same carrier content. To avoid these drawbacks, scholars have As shown in Fig. 2, firstly, bacterial (B. cohnii) spore suspension
immobilized nutrients and microorganisms in different groups of was centrifuged at 5000 rpm for 20 min. The bacterial spore sludge
carriers [7,16,23]. However, one problem caused by this method was suspended by distilled water and diluted to the concentration
is that the amount of carrier will increase significantly, resulting of bacterial spore solution of about 3.9  109 cell/cm3. The diluted
in contradicting effects on the self-healing and mechanical proper- bacterial suspension was adsorbed in the pore of EP particles under
ties of concrete cracks. vacuum pressure of 0.06 MPa for 15 min. After adsorption, EP
In summary, how to increase the concentration of bacteria and particles adsorbed by microbial spore solution was dried to a con-
nutrients in the crack area is one of the key issues to improve the stant weight at 40 °C. The number of microbial spores immobilized
crack-healing effect based on microbial mineralization while on perlite per milliliter was 1.1  109. Secondly, the mixed nutrient
 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222 3

Fig. 2. Process routing of self-healing agent.

solution of yeast extract, inosine, and L-calcium lactate (yeast cizer, and EP particles. Among them, the specimens without bacte-
extract, inosine, and L-calcium lactate concentrations were 1.5 g/L, rial spores were named ‘‘Control”. Bacterial spores immobilized on
2.68 g/L, and 20.8 g/L, respectively) was sprayed on the surface EP particles without any protective coating were directly added to
of EP particles using a watering can. The sprayed amount was the concrete mixture (namely N-B). In the G-B, C-B, S-B, H-B, and
180 g of this solution per liter of EP particles with bacterial spore. K-B, bacterial spores immobilized on EP particles (wrapped with
After spraying, EP particles were dried at 40 °C for 48 h to reach GP, PC, SC, HG, and MKPC, respectively) were added directly to
constant weight. Finally, we sprayed a protective shell on the sur- the concrete mixture. The apparent density of concrete mixtures
face of the particles using a high-pressure nozzle sprayer. In this can be measured according to the standard for test method of per-
study, the geopolymers (GP), Portland cement (PC), acidic sulfoalu- formance on ordinary fresh concrete [25]. Cylindrical specimens
minate cement (SC), potassium magnesium phosphate cement with dimensions of 100 mm  50 mm and 150 mm  40 mm
(magnesium oxide and potassium dihydrogen phosphate (KH2PO4), (d  h) were cast for each mixture. The specimens with the size
namely MKPC), and hemihydrate gypsum (HG) were selected as of 100 mm  50 mm were used for crack surface observation and
protective shells for EP particles, and the recipes of coatings are crack width measurement, and the specimens with the size of
shown in Table 1. After being packaged, the coated EP particles 150 mm  40 mm were used for water permeability experiment.
were cured in plastic bags for 12 h to ensure enough moisture dur- After 24 h, the concrete specimens were demoulded and then put
ing the parcel layer hardening, put into the oven at 40 °C, and then into standard curing box for 28 d. Cracks on all specimens were
dried until a constant weight was obtained. The resulting coated fabricated by the splitting method. The splitting specimen was
particles were directly added to the concrete mixture. tightened by a self-locking metal band. Before the binding, silica
gel films with different thicknesses were embedded in the cracks
2.3. Mix proportions and specimen preparation to form cracks with different widths. Finally, the specimens were
marked and immersed horizontally in tap water in a plastic bucket
As shown in Table 2, seven different types of concrete speci- for crack healing.
mens were prepared with coarse aggregate (diameter 5–16 mm),
natural sand, cement (P.O. 42.5), silica fume, water, superplasti- 2.4. Analytical methods

2.4.1. Surface observation method

Table 1
Mixtures used to coat EP particles. After crack-healing for 0 d, 7 d, 14 d, and 28 d, the surface of the
cracks was observed by the KLN-CBY501 image microscope, and
Item GP PC SC MKPC HG
the apparent morphology of the cracks was recorded after the
EP (g) 200 200 200 200 200 repair of the cracks. The width of the cracks was measured by
Water (g) 300 450 450 450 450
CAM-MS software. Basic principle of width measurement is that
Styrene-acrylate emulsion (g) 30 30 30 30 30
Metakaolin (g) 300 – – – – Micrometer and concrete cracks were observed under the same
Soluble silicate (g) 150 – – – – magnification, and the photographs were imported into CAM-MS
Cement (g) – 300 300 – – software. The Micrometer photos were used to set the scale. The
Magnesium oxide (g) – – – 75 – crack width can be measured by the parallel line spacing method
KH2PO4 – – – 225 –
after the setting is completed. Through the analysis of the recorded
Gypsum – – – – 300
photographs, the more mineral deposits formed in the fractured
4 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222

Table 2
Mix proportions of the concrete specimens.

Item Control N-B G-B C-B S-B H-B K-B

3
Coarse aggregate (kg/m ) 1211 1211 1211 1211 1211 1211 1211
Sand (kg/m3) 519 519 519 519 519 519 519
Cement (kg/m3) 484 484 484 484 484 484 484
Water (kg/m3) 290 290 290 290 290 290 290
Silica fume (kg/m3) 36 36 36 36 36 36 36
Superplasticizer (kg/m3) 5.2 5.2 5.2 5.2 5.2 5.2 5.2
Expanded perlite 0.5 m3 0.5 m3 0.5 m3 0.5 m3 0.5 m3 0.5 m3 0.5 m3
Coating None None GP PC SC HG MKPC
Apparent density (kg/m3) 2326 2322 2136 2145 2152 2197 2087

region after the same healing time, the better the reparability of 2.4.3. Width repair rate method
the self-healing concrete specimens. Generally, crack width is different along the direction of crack
length. Therefore, comparing the average width of cracks is not
2.4.2. Crack-healing width distribution method reasonable. To evaluate the crack-healing ability of each concrete
After crack-healing for 28 d, the widths of completely repaired specimen, we can measure the width of the crack over time and
cracks in each group were measured, and the distribution of crack find out the repair rate to characterize the repair effect. The
widths was analyzed. The wider the distribution range, the greater crack-healing rate under a certain repair time was calculated as
the width of repair, and the better the healing effect. follows:

Fig. 3. Water permeability experiment process.

Fig. 4. Water permeability setup.

 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222 5

Table 3
Comparison of unwrapped and wrapped expanded perlite.

Group None GP PC SC HG MKPC

Particle size (mm) 2.36–4.75 2.36–4.75 2.36–4.75 2.36–4.75 2.36–4.75 2.36–4.75
Bulk density (kg/m3) 78.9 190.6 187.2 193.7 189.4 191.5
Numerical tube Pressure (kPa) 152.5 452.2 411.6 432.5 387.6 547.6
Water absorption of 24 h (wt/%) 275 32 35 28 26 16

0 day 3 days 7 days 28 days

A
0.25

1mm

B
0.27

1mm

C
0.39

1mm

0.4
1mm

E
0.79

1mm

0.76

1mm

G
1.24

1mm

Fig. 5. Crack-healing process of concrete: control (A), N-B (B), G-B (C), P-B (D), S-B (E), H-B (F), and K-B (G).
6 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222

Healing percentage ð%Þ ¼ ðd0  dt Þ=d0  100 ð1Þ 3.2. Self-healing characterization

where d0 is the initial crack width, and dt is the width measured at To evaluate the MICP performance of the coating mat, we pre-
healing time t. pared seven types of concrete specimens to characterize their
The width repair rate method can directly reflect the mineral- self-healing ability. Fig. 5 shows the measured points of seven
ization of the surface area of the crack, and it is more convenient groups of specimens with cracks after healing times of 0, 7, 14,
for data acquisition and treatment since the results are more accu- and 28 d. The crack width of concrete specimens with self-
rate and the application scope is more extensive. healing agent decreased gradually with the increase of healing
time, and MICP crystals can be observed on their surface. The
cracks of concrete specimens were almost completely healed after
2.4.4. Water permeability experiments 28 d of healing, and the crack width ranges from 0.27 mm to
After crack-healing for 0 d, 3 d, 7 d, 14 d, and 28 d, the water 1.24 mm. The cracks in the control group were not completely
permeability experiment was carried out. In order to avoid the repaired, and the width of the crack was 0.25 mm. The reason is
influence of nutrient loss on the test results in the process of seep- that concrete can have a self-healing ability by itself due to the
age, the same batch but not the same specimen was used in the hydration reaction between unhydrated cement particles and
water permeability experiment under different healing times. external water, and the carbonation process of calcium hydroxide
Specimens with dimensions of 150 mm  40 mm were cast for (Ca(OH)2) in the smaller widths of cracks [27].
each mixture for water permeability experiments (as shown in Fig. 6 shows the crack width distribution of six groups of spec-
Fig. 3). The water permeability setup is shown in Fig. 4 [26]. The imens, N-B, G-B, P-B, S-B, H-B, and K-B, after being repaired for
permeability coefficient of each specimen at different healing 28 d. It can be seen from the diagram that after 28 d of healing,
times was calculated as follows: EP wrapped with GP, PC, SC, HG, and MKPC in concrete specimens
had good repairing effect for cracks with widths below 0.39, 0.4,
k ¼ 1500=ðlav e  wav e  t Þ ð2Þ 0.79, 0.76, and 1.24 mm, respectively. The crack repair width of
concrete specimens was less than 0.27 mm when the expanded
where k is the permeability coefficient (cm/s); lave is the average perlite was not wrapped. Expanded perlite wrapped with SC, HG,
length (cm) of cracks on both sides of the specimen; wave is the and MKPC was mixed with concrete, and the specimen had a
average width (cm) of cracks (found by measuring the width once higher crack repair width due to the lower pH values of sulfoalumi-
per cm along the crack length direction and then taken the average nate cement, gypsum, and potassium magnesium phosphate
value, wave = (w1 + w2 + . . . + wn)/n); t is the time spent on 1500 mL cement (10.2, 8.5, and 7.6, respectively), thereby prolonging the
water consumption in the permeability test. The smaller the perme- metabolic activity of bacteria.
ability coefficient of the specimen, the better the self-healing effect Fig. 7 shows the relationship between the initial crack width
of the crack. and crack-healing percentage for the six types of specimens. 20
Besides, for the specimens with better repair effect for cracks, measurements were made on each sample. As can be seen from
there will be some water remaining in the container during the Fig. 7A and B, when the initial crack width was 0.1–0.25 mm, the
seepage test, and the specimens will no longer leak. The permeabil- cracks in S-B, H-B, and K-B were almost completely healed (espe-
ity coefficient of specimen was zero at this time. The water head cially in the early stage of healing). Moreover, only a few cracks
height of seepage control can be calculated as follows: in N-B, G-B, and P-B have been completely healed in less than
7 d. The crack-healing percentages were increased as healing times
h ¼ v =pr2 ð3Þ were prolonged (Fig. 7C). When the initial crack width was 0.25–
0.65 mm, the lower crack-healing percentage was observed in N-
where h is the water head height (cm), v is the volume of the B, G-B, and P-B, while S-B, H-B, and K-B were almost completely
remaining water (cm3), and r is the diameter of the specimen (cm). repaired. The crack-healing effect of K-B was the best. Moreover,
as shown in Fig. 7C, at the same repair time, the repair effects of
G-B, P-B, S-B, H-B, and K-B were more remarkable than that of
3. Results and discussion N-B. The reason for the decrease of self-healing ability of N-B
was the unwrapped perlite, which reduced the germination rate
3.1. Properties of expanded perlite

The bulk density, the numerical tube pressure values, and water
absorptions of EP particles wrapped with various materials were
measured according to the standard method for lightweight aggre-
gates [28]. As shown in Table 3, the bulk density of the EP wrapped
with GP, PC, SC, HG, and MKPC were 2.41, 2.37, 2.46, 2.4, and 2.43
times that of the unwrapped EP, respectively. The bulk density of
the EP after being wrapped was significantly improved, and the
increase of the density of the carrier was favorable to the uniform
distribution of the carrier in the concrete. The numerical tube pres-
sure values of the EP wrapped with GP, PC, SC, HG, and MKPC were
2.97, 2.70, 2.84, 2.54, and 3.59 times that of the unwrapped EP.
Increased tube pressure was beneficial to increasing the strength
of the concrete. At the same time, the damage rate of EP in the con-
crete mixing process was reduced. The water absorption values of
the EP wrapped with GP, PC, SC, HG, and MKPC were reduced by
88.4%, 87.3%, 89.9%, 90.5%, and 94.1%, respectively. The decrease
of water absorption of perlite was beneficial to prevent the recov-
ery of microbial spores in the concrete-mixing process. Fig. 6. Width distribution of crack healing for 28 d.
 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222 7

Fig. 7. The relationship between the initial crack width and crack-healing percentage for the six types of specimens at healing times of 3 days (A), 7 days (B), and 28 days (C).
The relationship between the maximum repaired width and healing time (D). Permeability coefficients of specimens at different healing durations (E).

of bacterial spores in a high-alkaline environment. On the other crack-healing ability. Especially, the specimen K-B’s completely
hand, the strength of EP without wrapping was lower, and the healed width reached 1.24 mm after 28 d of healing, while the val-
breakage rate of EP carriers was higher during concrete mixing, ues observed in S-B and H-B were 0.79 and 0.76 mm, respectively.
leading to the decrease of the number of bacterial spores per cubic Therefore, the self-healing technique using immobilized bacterial
concrete. Moreover, the lower crack-healing percentage and heal- spores in EP wrapped with low-alkaline materials could improve
ing width of the G-B and P-B were attributed to the high alkalinity the crack-healing ability of concrete compared with that of EP
of the wrapping materials. wrapped with high-alkaline materials such as GP and PC.
Fig. 7D shows the relationship between the maximum repaired
width and healing time for the six types of specimens. As can be 3.3. Water permeability of specimens
seen from the diagram, with the increase of healing time, the heal-
ing widths of specimens N-B, G-B, and C-B did not increase Fig. 7E shows permeability coefficients of specimens for differ-
significantly. In contrast, S-B, H-B, and K-B exhibited much better ent healing times. As shown in the diagram, with the increase of
8 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222

the healing time, the permeability coefficients of all specimens

decreased gradually. Especially when the healing time reached
Calcite
14 d, the specimens S-B, H-B, and K-B could withstand water pres-
sure of 0.35, 0.65, and 1.1 cm head height, respectively. After 28 d
of healing, the water head hight of specimens S-B, H-B, and K-B
were 0.7, 1.1, and 1.7 cm, respectively. However, the seepage stop
was not observed in N-B, G-B, and C-B specimens. Therefore, the
self-healing technique using immobilized bacterial spores in EP
wrapped with low-alkaline materials could improve the imperme-
ability of specimens at cracks, thereby conforming to the findings
of microscopic observations.

3.4. Microstructure of self-healing agent Fig. 9. X-ray diffraction of the precipitations at the crack of K-B.

Fig. 8A shows the FESEM image of an EP particle. It can be seen

that there are a large number of micropores in EP. Therefore, it is
feasible to use EP particles as a carrier for bacterial spores. of EP particles. Fig. 8E shows that spraying a protective shell on the
Fig. 8C and D show the morphology of bacterial spores in the cavity surface of the self-healing agent led to a compact and uniform
of EP particles. A large number of bacterial spores were immobi- structure using a high-pressure nozzle sprayer, and the protective
lized in the cavities of the EP particles. These spores agglomerated shell could thus provide effective protection for the self-healing
and adhered to each other in the cavity and distribute unevenly, agent.
which was also the reason for different repair rates along the same Fig. 8F shows the microstructure of crack-healing precipitations
crack. Fig. 8B shows the micromorphology of the L-calcium lactate, of K-B. The X-ray diffraction analysis showed that precipitations at
which was used as a source of calcium during MICP, on the surface the crack of the specimen reached the highest peak at a 2h value of
29.42°, which corresponds to pure calcite (Fig. 9). The chemical
process that formed this CaCO3 can be attributed to the degrada-
A B tion of organic compounds into CO2 and H2O by B. chonii. In a
high-pH environment, CO2 can be easily converted into CO2 3 , and
in the presence of Ca2+, CaCO3 can be formed. In this case, the L-
calcium lactate was used as the organic carbon source. Therefore,
the overall biochemical reaction can be given as:
CaðC3 H5 O3 Þ2 þ 6O2 ! CaCO3 þ 5CO2 þ 5H2 O ð4Þ
Calcium lactate In addition, CO2 produced by Eq. (4) can also react with Ca(OH)2
in the concrete matrix to produce CaCO3 (Eq. (5)) and is the reason
why the control specimens have certain self-repairing ability (CO2
in water reacts with Ca(OH)2 in the concrete matrix).
CO2 þ CaðOHÞ2 ! CaCO3 þ H2 O ð5Þ
C D
4. Conclusion
B. cohnii spores
EP particles wrapped with different materials as novel bacterial
carriers and their effects on the crack-healing capacity of concrete
B. cohnii spores have been investigated. The following results were obtained:

(1) After the EP particles were coated, its numerical tube pres-
sure was increased, and the water absorption was reduced.
Wrapping materials can effectively reduce the breakage rate
of EP particles and resist the invasion of water in the process
E F of mixing concrete.
(2) The EP particles after the wrapping process were found to
improve the healing capacity of concrete cracks. When EP
particles immobilized with bacterial spores were wrapped
with a low-alkaline material, the crack-healing effect of con-
crete was the best. The completely healed width reached
1.24 mm after 28 d of healing.
(3) Water permeability experiments showed that the self-
healing technique using immobilized bacterial spores in EP
wrapped with low-alkaline materials could improve the
impermeability of specimens at cracks. When the healing
time reached 14 d, the specimens S-B, H-B, and K-B with-
Fig. 8. The FESEM images of EP particles (A) with L-calcium lactate on the surface
stood water pressure of 0.35, 0.65, and 1.1 cm head height,
(B), after coating B. cohnii spores (C and D) with the protective shell of K-B (E). The while no impermeability was observed in the samples with
microstructure of crack-healing precipitations of K-B (F). EP wrapped with high-alkaline materials.
 L. Jiang et al. / Construction and Building Materials 232 (2020) 117222 9

(4) The X-ray diffraction analysis showed that the precipitated [10] W. De Muynck, N. De Belie, W. Verstraete, Microbial carbonate precipitation in
construction materials: a review, Ecol. Eng. 36 (2) (2010) 118–136.
mineral on the crack surface was calcite. Microstructure
[11] J.L. Zhang, R.S. Wu, Y.M. Li, et al., Screening of bacteria for self-healing of
analysis confirmed that a large number of bacterial spores concrete cracks and optimization of the microbial calcium precipitation
were immobilized by the EP particles; thus, EP has mani- process, Appl. Microbiol. Biot. 100 (15) (2016) 6661–6670.
fested itself as an excellent bacterial carrier. [12] J. Zhang, B. Mai, T. Cai, et al., Optimization of a binary concrete crack self-
Healing system containing bacteria and oxygen, Materials 10 (2) (2017) 116.
[13] M. Luo, C. Qian, R. Li, Factors affecting crack repairing capacity of bacteria-
based self-healing concrete, Constr. Build. Mater. 87 (2015) 1–7.
Declaration of Competing Interest [14] J. Wang, J. Dewanckele, V. Cnudde, et al., X-ray computed tomography proof of
bacterial-based self-healing in concrete, Cem. Concr. Comp. 53 (2014) 289–
304.
The authors declare that they have no known competing finan- [15] J.Y. Wang, H. Soens, W. Verstraete, et al., Self-healing concrete by use of
cial interests or personal relationships that could have appeared microencapsulated bacterial spores, Cem. Concr. Res. 56 (2014) 139–152.
[16] M. Alazhari, T. Sharma, A. Heath, et al., Application of expanded perlite
to influence the work reported in this paper. encapsulated bacteria and growth media for self-healing concrete, Constr.
Build. Mater. 160 (2018) 610–619.
Acknowledgments [17] J.Y. Wang, N. De Belie, W. Verstraete, Diatomaceous earth as a protective
vehicle for bacteria applied for self-healing concrete, J. Ind. Microbiol.
Biotechnol. 39 (4) (2012) 567–577.
Funding was provided by China Postdoctoral Science Founda- [18] J. Zhang, Y. Liu, T. Feng, et al., Immobilizing bacteria in expanded perlite for the
tion (No. 2018M631776), National Science Foundation of China crack self-healing in concrete, Constr. Build. Mater. 148 (2017) 610–617.
[19] W. Khaliq, M.B. Ehsan, Crack healing in concrete using various bio influenced
(Nos. 51708385, 51678384, and 51808375), Natural Science Foun-
self-healing techniques, Constr. Build. Mater. 102 (1) (2016) 349–357.
dation of Shanxi Province, China (No. 201801D121095), and Scien- [20] N.T.H. Nguyen, M.P. Nghi, P.A.T. Nguyen, et al., Bacillus subtilis HU58
tific and Technological Innovation Programs of Higher Education immobilized in micropores of diatomite for using in self-healing concrete,
Institutions in Shanxi (No. 2019L0324). in: T.C. Tim, T. Ueda, and H.S. Mueller, (Eds.), Procedia Engineering, (2017)
598–605.
[21] E. Tziviloglou, V. Wiktor, H.M. Jonkers, et al., Bacteria-based self-healing
References concrete to increase liquid tightness of cracks, Constr. Build. Mater. 122 (2016)
118–125.
[1] N.Z. Muhammad, A. Shafaghat, A. Keyvanfar, et al., Tests and methods of [22] E. Tziviloglou, Z. Pan, H.M. Jonkers, et al., Bio-based self-healing mortar: an
evaluating the self-healing efficiency of concrete: a review, Constr. Build. experimental and numerical study, J. Adv. Concr. Technol. 15 (9) (2017) 536–
Mater. 112 (2016) 1123–1132. 543.
[2] D. Snoeck, P. Van den Heede, T. Van Mullem, et al., Water penetration through [23] J. Wang, K. Van Tittelboom, N. De Belie, et al., Use of silica gel or polyurethane
cracks in self-healing cementitious materials with superabsorbent polymers immobilized bacteria for self-healing concrete, Constr. Build. Mater. 26 (1)
studied by neutron radiography, Cem. Concr. Res. 113 (2018) 86–98. (2012) 532–540.
[3] M.G. Sierra-Beltran, H.M. Jonkers, E. Schlangen, Characterization of sustainable [24] G.T. 7476-1987, Water Quality-Determination of Calcium – EDTA titrimetric
bio-based mortar for concrete repair, Constr. Build. Mater. 67 (2014) 344–352. method, China Environmental Science Press, Beijing, China, 1987 (Only
[4] V. Wiktor, H.M. Jonkers, Quantification of crack-healing in novel bacteria- available in Chinese).
based self-healing concrete, Cem. Concr. Comp. 33 (7) (2011) 763–770. [25] G.T. 50080-2002, Standard for Test Method of Performance on Ordinary Fresh
[5] H.M. Jonkers, A. Thijssen, G. Muyzer, et al., Application of bacteria as self- Concrete, China Architecture & Building Press, Beijing, China, 2002 (Only
healing agent for the development of sustainable concrete, Ecol. Eng. 36 (2) available in Chinese).
(2010) 230–235. [26] J. Zhang, C. Zhao, A. Zhou, et al., Aragonite formation induced by open cultures
[6] M. Wu, B. Johannesson, M. Geiker, A review: self-healing in cementitious of microbial consortia to heal cracks in concrete: insights into healing
materials and engineered cementitious composite as a self-healing material, mechanisms and crystal polymorphs, Constr. Build. Mater. 224 (2019) 815–
Constr. Build. Mater. 28 (1) (2012) 571–583. 822.
[7] H. Chen, C. Qian, H. Huang, Self-healing cementitious materials based on [27] C. Edvardsen, Water permeability and autogenous healing of cracks in
bacteria and nutrients immobilized respectively, Constr. Build. Mater. 126 concrete, ACI Mater. J. 96 (4) (1999) 448–454.
(2016) 297–303. [28] G.T. 17431.2-2010, Lightweight Aggregates and Its Test Methods-Part 2. Test
[8] G. Souradeep, P.S. Dai, K.H. Wei, Autonomous healing in concrete by bio-based Methods for Lightweight Aggregates, China Environmental Science Press,
healing agents - A review, Constr. Build. Mater. 146 (2017) 419–428. Beijing, China, 2010 (Only available in Chinese).
[9] K. Vijay, M. Murmu, S.V. Deo, Bacteria based self healing concrete - A review,
Constr. Build. Mater. 152 (2017) 1008–1014.
Annotations

Sugar-coated expanded perlite as a bacterial carrier for

crack-healing concrete applications
Jiang, Lu; Jia, Guanhua; Jiang, Chen; Li, Zhu

01 Camilo Andrés Agudelo Gallo Page 1

13/4/2020 13:41

02 Camilo Andrés Agudelo Gallo Page 1

13/4/2020 13:41

03 Camilo Andrés Agudelo Gallo Page 1

13/4/2020 13:41

04 Camilo Andrés Agudelo Gallo Page 1

13/4/2020 13:41

05 Camilo Andrés Agudelo Gallo Page 1

13/4/2020 13:43

06 Camilo Andrés Agudelo Gallo Page 1

13/4/2020 13:43

07 Camilo Andrés Agudelo Gallo Page 1

13/4/2020 13:44