Journal of Building Engineering 98 (2024) 111269

Contents lists available at ScienceDirect

Journal of Building Engineering

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

Rebar corrosion and bond in concrete with recycled glass granules

Liangwen Wei a,b , Nan Zhang c,*, Liang Sun b
a
State Key Laboratory of Mountain Bridge and Tunnel Engineering, Chongqing Jiaotong University, Chongqing, China
b
School of Civil Engineering, Chongqing Jiaotong University, Chongqing, China
c
Centre for Sustainable Infrastructure and Digital Construction, School of Engineering, Swinburne University of Technology, Victoria, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Waste glass granules are considered a cost-effective alternative material to replace aggregates in
Recycled glass granules concrete. However, the bond degradation of corroded concrete containing recycled glass granules
Concrete has not been explored. This study investigated the effect of glass granules on rebar corrosion and
Rebar corrosion rate
the bond loss between rebars and concrete subjected to accelerating corrosion. The compressive
Bond strength loss
Pore structure
strength and pore structure of the concrete were first evaluated, and the corrosion rate and pull-
out testing of rebar in concrete were conducted after corrosion. The results show that, despite
reducing the initial bond strength before corrosion, a high replacement of recycled glass granules
improved the concrete’s resistance to corrosion due to the optimized microstructure of concrete.
In addition, the compressive strength of the concrete dominated the initial bond strength, which
reduced with increasing the corrosion period. However, this negative effect of corrosion was
mitigated by increasing the recycled glass granules, followed by reduced bond strength loss and a
lower rebar corrosion rate.

1. Introduction

In general, the bond between reinforcing bars and concrete is essential in reinforcement concrete structures, such as bearings
transmitting concentrated loads, beam-column joints in buildings, and rock bolts used for rock support in tunnel engineering, expe­
riencing stress concentration, shear, and contra flexure [1–5]. In terms of the bond between reinforcing bars and concrete, there are
three main factors, including chemical adhesion, mechanical interlocking action, and friction, which are dependent on the bleeding
and casting direction of fresh concrete, the strength of hardened concrete, the rib geometry of reinforcing bars, and the confining stress
at the reinforcing bar-concrete interface [5]. According to the Model Code MC 2010 [6], a typical bond stress-slip curve has three
stages, as shown in Fig. 1. Stage I is an ascending branch with high initial bond stiffness. Bond stiffness reflects the increasing rate of
bond stress. The bond stress will reduce with the increase of micro-cracking in the surrounding concrete and the breakdown of
chemical adhesion between concrete and rebar. Stage II is a sustained plateau with nearly constant bond stress. If the surrounding
concrete is well-confined, the condition will occur due to the plasticity provided by the confinement. Stage III is a descending branch
due to bond failure by either bar slipping or concrete splitting during the slip process. However, these reinforcement concrete
structures are commonly exposed to chloride ions in practical applications, which will inevitably cause inadequate bonding.
To date, the interfacial bonding performance between rebar and concrete under corrosion conditions has been widely investigated
in the past years. As chloride ions can quickly disrupt the protective passivation film on steel reinforcements, the rebar corrosion is

* Corresponding author.
E-mail address: [email protected] (N. Zhang).

https://doi.org/10.1016/j.jobe.2024.111269
Received 10 July 2024; Received in revised form 11 October 2024; Accepted 5 November 2024
Available online 6 November 2024
2352-7102/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

accelerated. This not only deteriorates the material properties of the steel bars, but also weakens the chemical adhesion, mechanical
interlocking action, and friction between the steel bars and the concrete, and even causes the concrete to crack [7]. As a result, the
reinforcing bars could cause excessive bond slip while in service, followed by ineffective anchorage [8]. This phenomenon has
occurred in actual projects. Considering the lengthy duration required for natural corrosion tests, previous studies widely used the
electrochemical accelerated corrosion method as the evaluation method [9]. By utilizing empirical models and theoretical analyses,
Lin and Zhao [10] even developed a degradation model describing the deterioration of interfacial bonding properties due to corrosion.
However, these previous studies mainly concentrated on the degradation of ordinary concrete with river sand [11].
In fact, due to rapid economic development and extensive infrastructure projects, the escalating demand for river sand has become
a significant environmental concern over the past two decades. For instance, the construction of a single kilometer of highway requires
approximately 30 tons of river sand, while a nuclear power plant might need as much as 12 million tons [12]. In addition, global sand
and gravel consumption has tripled due to large-scale and often unrestricted extraction of river sand resources. This surge in demand
has not only led to a depletion of natural river sand but also contributed to severe environmental issues [13–15]. In this context, waste
glass granules have started to be reused as a cost-effective alternative material to sand and coarse aggregates in concrete [16,17]. On
the one hand, their accumulation has become a pressing environmental issue with the proliferation of glass products in daily life [18].
On the other hand, most waste glass ends up in landfills, causing environmental pollution and resource wastage. Therefore, the
degradation of concrete containing coarse glass granules should be explored.
Until now, previous studies have investigated the fresh, hardened, and durability properties of concrete mixed with recycled glass.
For example, Bisht and Ramana [19] evaluated the workability, compressive strength, flexural strength, density, and water absorption
of concrete by substituting sand with fine glass aggregates at varying percentages from 18 % to 24 %. It was found that the substitution
of 21 % was optimal to enhance the microstructure and mechanical properties of concrete but the increase of recycled glass content
gradually reduced the water permeability. Su and Xu [20] studied the effect of recycled glass sand and rice husk ash on concrete and
found that adding glass sand reduced the water absorption and compressive strength of concrete. It, however, improved the split
tensile strength, shear strength, and cyclic compression properties. Besides, Małek et al. [21] also investigated the mechanical
properties of concrete containing recycled glass and substituted fine aggregates with different contents of 5 %, 10 %, 15 %, and 20 % by
mass. The results showed that the compressive and flexural strength of concrete were enhanced by adding glass sand aggregates, but
the increase of split tensile strength was minimized. Additionally, Kim et al. [22] found that adding cathode ray tube waste glass as fine
aggregates reduced compressive and flexural strengths but highlighted that the resistance to freezing and thawing, sulfate attack, and
chloride penetration was improved. However, rebar corrosion and bonding degradation in concrete with recycled glass granules
during corrosion have never been reported.
This study aims to investigate the effect of recycled glass granules on rebar corrosion and bond loss in corroded concrete. Recycled
glass particles are selected to replace natural river sand with similar particle sizes. First, the compressive strength and pore structure of
concrete containing different fractions of recycled glass sand are tested to understand the properties of concrete. Secondly, the
corrosion rate of steel bars in concrete containing recycled glass granules is evaluated to figure out the key impact factors of concrete
on rebar corrosion. Finally, the change in bond strength between rebar and concrete with varying corrosion periods was compared.

2. Materials and experimental methods

2.1. Rebars

This study used a type of threaded steel rebar with a diameter of 6 mm. The mechanical properties are shown in Table 1. Before
casting specimens, these reinforcing bars were first cleaned by acid pickling to remove rust and then air-dried for preparation. The

Fig. 1. Typical bond stress-slip curve [5].

2
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

initial mass of the rebars was recorded as m0.

2.2. Cement

P.O. 42.5 cement produced by Neoshui Red Lion Cement Co. was used in this study. Its specific density and surface area were 3060
kg/m3 and 334 m2/kg, respectively. The initial time and final setting times were measured at 212 and 281 min.

2.3. Sand

Two types of sand, including river sand and glass sand, were used with comparable particle size distributions, as shown in Fig. 2.
The glass granules were obtained by washing, crushing, and sieving the recycled glass raw materials in glassware factories. The water
absorption rate of sand was between 2 % and 6 %.

2.4. Mixture proportions

Table 2 shows the mixture proportions of concrete, in which the water-to-cement ratio was 0.5 and the cement-to-sand ratio
remained at 1. The river sand was gradually replaced with glass granules of up to 100 % substitution rates.

2.5. Mixing process

Cement powder and water were first mixed in a mixer at a low speed of 62 RPM for 60 s. During the last 30 s of the mixing, the sand
mixture was gradually added to the mixer. Thereafter, the mixing speed increased to 125 RPM, and the mixture was mixed for another
30 s. Then, the mixer was paused for 90 s to scrape the remaining concrete off the inner walls of the container. Finally, the concrete was
mixed for a further 60 s to obtain the final mixture.

2.6. Experimental procedures

2.6.1. Strength testing

The compressive strength of concrete was tested according to BS EN 196 1:1995 [23]. The fresh concrete was poured into the oiled
molds and vibrated for 1 min to remove air bubbles from the mixture. After curing for 24 h, the samples sealed with plastic wrap were
demolded and placed in a controlled environment at a temperature of 20 ± 2 ◦ C and a humidity of 90 ± 5 ◦ C until testing. The
specimens were used for compression testing between two 40 mm × 40 mm square plates at a load rate of 0.33 MPa/s. Six samples were
tested to obtain the average.

2.6.2. Mercury intrusion porosimetry (MIP)

A Micromeritics® automated mercury piezometer (9600 AutoPore V) was used to evaluate the pore structure of specimens. Small
pieces of samples were first collected from broken specimens for strength and then dried at 105 ◦ C for 24 h before cooling at room
temperature. The mercury pressure was controlled at 33,000 psi and the contact angle between the mercury and the samples was 130◦
during testing.

2.6.3. Pull-out testing

To evaluate the bonding between the reinforcement and concrete, the rebar was inserted in the center of the cylindrical concrete
specimen with a diameter of 50 mm and a height of 100 mm. Due to non-uniform bond stress along the bond length of the steel bar, a
bond length that is too long may underestimate the bond stress, while a bond length that is too short may cause a large dispersion in the
test results. Additionally, the accelerated corrosion was expected significantly to reduce the bond behaviors. To avoid invalid com­
parison tests due to an inadequate bond length, this study designed a bonding length of 80 mm, which is around 13 times the diameter
of the rebar, as shown in Fig. 3.
The interfacial bond strength was tested by the customized setup, as shown in Fig. 4. The rebar-concrete sample was placed into the
reaction frame device and a steel mold was used to provide annular confinement around the specimen. The reaction frame was fixed to
the upper clamp of the testing machine while the lower clamp was connected to the exposed rebar. The loading rate was set to 0.5 mm/
min. Based on the pull-out failure load, the bond strength τ was calculated by Eq. (1) [8]:
F
τ= (1)
π dl

Table 1
Mechanical properties of the threaded steel rebar used in this study.
Rebar type Nominal diameter (mm) Yield strength (MPa) Tensile strength (MPa)

HRB335 6 335 445

3
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

Fig. 2. Particle size distributions of river sand and recycled glass granules.

Table 2
Mixture proportions of concrete with different fractions of sand.
No. Replacement rate of glass granule Cement River sand(kg/m3) Glass granule(kg/m3) Water
(kg/m3) (kg/m3)

G0 0% 1600 1600 0 800

G25 25 % 1600 1200 400 800
G50 50 % 1600 800 800 800
G75 75 % 1600 400 1200 800
G100 100 % 1600 0 1600 800

Fig. 3. Dimensions of rebar-concrete specimens in this study.

where F is the failure load, d is the diameter of the reinforcing bar, and l is the bonded length of the reinforcing bar.

2.6.4. Accelerated corrosion procedure

Based on Faraday’s law, this study used electrochemically accelerated corrosion with a high current density to simulate the
corrosion process of steel in concrete, which has been widely employed in related studies. Fig. 5 presents the experiment apparatus of
the full immersion electrified corrosion method. To better simulate the accelerated corrosion of chloride ions, it is generally recom­
mended to set the concentration of chloride salt solution to twice the concentration of seawater, so the specimens in this study were
soaked in a 7 % NaCl electrolyte solution during the accelerated corrosion. The steel bar and the carbon rod as the anode and the
cathode of the primary cell were connected to the positive and negative poles of the DC power supply, respectively. After the circuit
was connected, the current output was maintained at a constant value of 2000 μA/cm2 . The change in resistance reflected the
development of corrosion.

4
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

Fig. 4. Schematic diagram of pull-out testing.

Fig. 5. Experiment apparatus for electrochemically accelerated corrosion.

2.6.5. Corrosion rate testing

After pull-out testing, the pulled rebars were used to evaluate the corrosion rate of reinforcement. The initial mass m0 of the
reinforcement had been recorded before casting. First, the concrete residue attached to the reinforcement was removed by acid
pickling and descaling. Next, the rebar was cleaned with the neutralizing solution of water and 3 % sodium carbonate solution in turn.
Finally, the reinforcement was dried and weighed to record the mass m1. The corrosion rate can be calculated by Eq. (2).
m0 − m1
ω= ⋅ 100% (2)
m0

3. Results and discussion

3.1. Properties of concrete

3.1.1. Compressive strength

Fig. 6 shows the compressive strength of concrete with recycled glass granules. The average data presented were obtained by
testing six samples. All The mixture without recycled glass granules was considered the control. The river sand was gradually replaced
with glass granules, up to a maximum of 100 %. As expected, the compressive strength of all mixtures increased with increasing ages.

5
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

For a given curing age, these specimens first exhibited an increase in compressive strength by replacing glass sand with 25 %. The
compressive strength at 3 days, 7 days, and 28 days reached 21.6 MPa, 32.1 MPa, and 40.2 MPa, respectively. A similar trend was also
found by Małek et al. [21] using fine glass aggregates to replace sand. This increase could be attributed to the pozzolanic reaction of
fine glass particles [24] and the refinement of pore structure [25]. With the further replacement of glass granules, the compressive
strength started decreasing, regardless of curing age. This could be due to the smooth surface of recycled glass which hinders the
adhesion to the cement paste and thereby reduces the compressive strength [22]. It is worth mentioning that even though the
replacement of glass granules reached 100 %, the reduction in compressive strength was slight compared to the control one and
remained within 5 MPa. Table 3 presents the relative compressive strength of concrete with recycled glass granules to the control one.
It was found that the negative effect of glass granules on strength was compensated by increasing the curing age, which could be
related to the pozzolanic reaction between glass and cement.

3.1.2. Pore structure

Fig. 7 shows the pore distribution of samples with different substitutions of recycled glass sand. It was found that pore diameters
smaller than 200 nm exhibited the most obvious changes. Additionally, the peak of pore distributions mainly concentrated from 60 nm
to 100 nm, regardless of replacement rates. Interestingly, increasing the substitution rate of recycled glass aggregates significantly
decreased the number of pores in the concrete. For example, the cumulative pore areas of concrete with 0 %, 25 %, 50 %, 75 %, and
100 % recycled glass sand were 15.726 m2/g, 14.818 m2/g, 13.144 m2/g, 12.565 m2/g, 12.061 m2/g, respectively. When the
replacement rate reached 100 %, the cumulative pore area was only 76.7 % of the control. This indicates that the introduction of glass
aggregate is beneficial for the optimization of concrete microstructure. This could be because the SiO2-rich glass aggregate reacted
with cement-hydrated Ca(OH)2 and formed additional calcium silicate hydrate (C-S-H gel), thereby densifying the pore structure of
concrete. As a result, the cumulative pore area of concrete presented a reduction with the increase of substitution rate.
Table 4 shows the porosity of concrete containing glass sand of up to 100 %. Increasing the substitution rate of glass sand reduced
the porosity of concrete. When the river sand was replaced with over 75 %, the relative porosity exhibited an obvious reduction. For
the replacement of 100 %, the relative value was reduced to 71 % of the control one. Therefore, it can be concluded that replacing river
sand with recycled glass granules can effectively reduce the porosity in concrete, especially for high replacement. Interestingly, the
finding was opposite to the results of compressive strength. The reason could be related to the strength of aggregates, but this is beyond
the scope of this paper and will be discussed in the next study.

3.2. Corrosion of rebars

3.2.1. Rust deterioration

Fig. 8 presents the deterioration of specimens undergoing accelerated corrosion for 1, 2, and 3 days. At the beginning of the
accelerated corrosion test, rust products continuously overflowed from the reinforcement-concrete interface and increased with the
duration of corrosion time. After accelerating corrosion for 50 h, the concrete occurred cracking regardless of adding recycled glass
granules.
Fig. 9 shows the resistance of concrete tested during the accelerated corrosion process. The resistance of all specimens rose with the
increasing corrosion time, which can be attributed to the accumulation of rust products on the boundary between the rebar and the
concrete. When concrete cracked, the resistance had a steep drop after a peak. Notedly, owing to lower porosity, the concrete
incorporating glass granules exhibited higher resistance and longer corrosion periods before the drop. For example, the resistance of
G100 at the peak was the highest with 2072 Ω while the resistances of G50, G25, and control were 1837 Ω, 1734 Ω, and 1374 Ω in

Fig. 6. Compressive strength of concrete with different fractions of recycled glass granules.

6
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

Table 3
Relative compressive strength of concrete with recycled glass granules.
Replacement rates Curing time

3 days 7 days 28 days

0% 100 % 100 % 100 %

25 % 101 % 108 % 110 %
50 % 92 % 90 % 95 %
75 % 83 % 81 % 91 %
100 % 85 % 83 % 88 %

Fig. 7. Effect of glass sand substitution rate on pore distribution of concrete.

Table 4
Porosity and relative values of concrete containing recycled glass granules with
different proportions.
No. Porosity (%) Relative value

G0 22.83 100 %
G25 21.47 94 %
G50 20.77 91 %
G75 18.85 83 %
G100 16.19 71 %

Fig. 8. Status of these samples subjected to different corrosion times.

7
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

order. Considering that concrete cracking can lead to strength failure, the duration of accelerated corrosion was set within 48 h in this
study.

3.2.2. Rebar corrosion rates

Fig. 10 shows the rebar corrosion rates in concrete experiencing accelerated corrosion for 16, 32, and 48 h. The magnitude of
corrosion rates reflects the resistance of concrete to chlorine ions. The more the amount of chloride ions migrated to the surface of the
reinforcement, the faster the rate of rebar corrosion. All samples presented an increase in corrosion rates with the extension of
corrosion time.
However, it is worth noting that the control concrete exhibited the highest corrosion rate; the rebar corrosion rate was reduced by
increasing the recycled glass fraction. This finding indicates that adding recycled glass granules can effectively diminish the corrosion
of rebar in concrete exposure to chlorine salts. This result was coincidence with the effect of glass granules on pore structure. This
means that pore structures of concrete play a dominant role in the corrosion process of rebar in concrete. Incorporating glass granules
can optimize multi-hazardous pores and hazardous pores in concrete, thereby slowing down rebar corrosion.

3.3. Bonding behaviors between rebars and concrete

Fig. 11 shows the bond stress–slip curves of different concrete mixtures experiencing accelerated corrosion. As expected, the bond
stress–slip relationship of all specimens initially exhibited a steep increase in Stage I. After the peak of bond strength, these curves
entered a softening Stage II. Up to the point of failure, the bond strength continued to decrease in Stage III due to reduced adhesion and
friction between rebars and concrete [6].
It is worth noting that the accelerated corrosion and the glass granule incorporation significantly changed the bond stress–slip
relationship. Regarding the control concrete without glass granules, the increasing rate of bond stress during Stage I had an obvious
reduction after accelerating corrosion for 16 h. This indicates that the corrosion process had damaged the boundary between rebars
and concrete and weakened their chemical adhesion. With the further increase of accelerated corrosion periods, the peak of bond
strength became lower. A similar phenomenon can be seen in the specimens with glass granules. However, as the content of glass
increased, this negative effect gradually weakened and even disappeared when the river sand was completely replaced by glass
granules. This can be attributed to optimized pore distributions.
In Stage III, the descent rate of bond strength was steep, and the descending period became short with the increasing corrosion
period, regardless of the content of glass granules. Notedly, the descent rate became slow and the descending period almost remained
constant when the glass granules were incorporated to 100 %. Since the descending process could be related to the friction between
rebars and concrete [5], it can be concluded that using glass granules can weaken the friction degradation of concrete at a fixed
corrosion period; however, this effect was slight for part replacement.
Table 5 shows the peak of bond strength and its relative values of concrete with varying replacement rates of recycled glass granules
to the control. Regarding the specimens without corrosion, there is an optimum replacement rate of glass particles. Incorporating
recycled glass granules with 25 % firstly increased the peak of bond strength and the highest peak reached 9.93 MPa. Afterward, the
strength peak gradually reduced as glass granules increased from 25 % to 100 %. This trend coincided with the result of compressive
strength. This indicates that the bond strength between rebar and concrete before corrosion depends on the compressive strength of the
concrete.
However, the corrosion of steel bars significantly deteriorated the bonding performance. For the control mixture, accelerating
corrosion for 16 h reduced the bond peak from 9.57 MPa to 6.56 MPa while it decreased the one of concrete with 100 % glass granules
from 6.58 MPa to 5.65 MPa. Even for the best concrete containing 25 % glass particles, the bond strength also decreased to 7.12 MPa.

Fig. 9. Resistance change of concrete with increasing the accelerated corrosion time.

8
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

Fig. 10. Corrosion rate of reinforcement in concrete with different fractions of recycled glass granules.

More importantly, the longer corrosion time resulted in a lower peak of bond strength. When the corrosion period was extended to 48
h, the bond peak of the control reduced to 3.64 MPa. Interestingly, the use of recycled glass granules slowed down the reduction in
bond strength. The relative bond strength of concrete gradually increased with the increase of recycled glass granules. It can be seen
from Table 5 that, after accelerating corrosion for 16 h, the relative value of concrete without glass had reached 69 % whereas the one
with 100 % glass granules remained as high as 86 %. The same phenomenon also occurred at the corrosion times of 32 and 48 h;
however, the decreasing trend slowed down, meaning that the early stage of accelerating corrosion played the dominant role in bond
loss. For the specimen accelerated corrosion for 32 h, the relative bond strength of the control dropped to 54 % while the one with 100
% glass granules reached 75 %. At this time, the two specimens had similar bond strength with about 5.14 MPa although their initial
bond strengths differed by 3 MPa. With the further extension of accelerating corrosion to 48 h, the reduction of bond strength
continued. The relative value of the control had fallen to 38 % whereas the one with 100 % glass granules reached around 73 %.
Notedly, the bond strength for 100 % glass replacement had been much higher than the one for the control. This indicates that the
resulting bond strength of corroded concrete is independent of the compressive strength of the concrete. It can be attributed to refining
the pore size by incorporating recycled glass granules. Therefore, it can be concluded that even though increasing glass content reduces
the initial bond strength, it can slow the corrosion process and prevent a significant degradation in bond strength.
Fig. 12 shows the relationship between rebar corrosion and bond strength loss. As expected, the loss of bond strength was pro­
portional to the rebar corrosion rate, regardless of the replacement rate of glass sands. It is worth mentioning that for the control
specimens without glass sand, the loss rate of bond strength remained almost constant with the increase of corrosion rates. In com­
parison, regarding the specimens with glass sand, the loss rate of bond strength gradually slowed down with increasing the rebar
corrosion rates, especially for accelerating corrosion over 32 h. This indicates that even though the corrosion rates of rebars were fixed,
the concrete containing recycled glass granules exhibited a lower bond strength loss than the one without glass particles. This was
mainly because the initial bond strength of the former was lower than the one of the control while the rest of the bond strength after
corrosion was higher. It can also be seen that, for the samples with glass granules, the early stage (within 16 h) of the corrosion process
had a higher impact on reducing bond strength than the late stage. It was because the rust produced on the reinforcement-concrete
interface at the beginning of corrosion, consequently weakening the chemical adhesion between rebars and concrete.
More importantly, for a given corrosion rate, a higher replacement with glass sand brought a lower bond strength loss, especially
for a replacement higher than 75 %, in which the porosity of concrete occurred an obvious reduction. This indicates that using a high
content of glass sand adequately optimized the internal structure of the boundary between the concrete and the rebar. Although the
compressive strength of concrete was reduced, the refined concrete-reinforcing bar interface can enhance the interface adhesion,
friction, and mechanical interlocking action. As a result, the bond strength loss exhibited a reduction. However, this reason is not clear,
which will be required for further study.

4. Conclusion

This study has investigated the effect of recycled glass granules on rebar corrosion and bond loss in concrete. The strength and
microstructure of the concrete containing different contents of recycled glass sand were evaluated. The corrosion of steel bars and the
bond properties of rebar with the concrete were tested. Based on experimental results and analysis, the conclusions can be drawn as
follows:

9
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

Fig. 11. Bond slip curves of reinforcement in concrete.

Table 5
Peak of bond strength and their relative values of concrete with different replacement rates of recycled glass granules.
Replacement rates of recycled glass granules Corrosion time

0h 16 h 32 h 48 h

0% 9.57 (100 %) 6.56 (69 %) 5.14 (54 %) 3.64 (38 %)

25 % 9.93 (100 %) 7.12 (72 %) 5.99 (60 %) 5.36 (54 %)
50 % 8.33 (100 %) 6.67 (80 %) 5.76 (69 %) 5.23 (63 %)
75 % 7.04 (100 %) 5.79 (82 %) 5.28 (75 %) 5.01 (71 %)
100 % 6.58 (100 %) 5.65 (86 %) 5.15 (78 %) 4.79 (73 %)

10
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

Fig. 12. Relationship between the rebar corrosion rate and the bond strength loss of concrete with different replacement rates of recycled
glass granules.

1) By replacing river sand with recycled glass granules, particularly at high replacement rates, the porosity of the concrete was
effectively reduced. This optimized microstructure directly improved the concrete’s resistance to corrosion and extended the
corrosion period of rebar, despite reducing the initial bond strength between the concrete and the rebar before corrosion.
2) The compressive strength of uncorroded concrete was the primary factor determining the initial bond strength. The bond strength
decreased due to accelerating corrosion, especially in the early stages within the first 16 h; however, this negative effect gradually
diminished with higher proportions of glass granules. When the replacement rate reached 75 %, the loss in bond strength was
notably reduced.
3) Regardless of glass sand replacement rates, the corrosion rate of rebars was proportional to the loss of bond strength.
4) Incorporating recycled glass sand refined the pore distribution of concrete, especially for smaller than 200 nm. High substitution
rates of recycled glass aggregates not only reduced the porosity in concrete but also decreased the overall pore size.

Future research could further explore the bonding performance between different types of rebar and concrete containing glass
granules. This will provide valuable insights for an analytical bond-slip constitutive model.

CRediT authorship contribution statement

Liangwen Wei: Writing – review & editing, Writing – original draft, Supervision, Investigation, Funding acquisition, Conceptu­
alization. Nan Zhang: Writing – review & editing, Writing – original draft, Supervision, Methodology, Conceptualization. Liang Sun:
Writing – original draft, Investigation, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

References

[1] R. Ludacer, The World Is Running Out of Sand-And There’sa Black Market for it now[J], The Business Insider, 2018.
[2] F. Colangelo, A. Petrillo, R. Cioffi, et al., Life cycle assessment of recycled concretes: a case study in southern Italy, Sci. Total Environ. 615 (2018) 1506–1517.
[3] C.J. Engelsen, H.A. van der Sloot, G. Petkovic, Long-term leaching from recycled concrete aggregates applied as sub-base material in road construction, Sci.
Total Environ. 587 (2017) 94–101.
[4] M. Wijayasundara, P. Mendis, R.H. Crawford, Integrated assessment of the use of recycled concrete aggregate replacing natural aggregate in structural concrete,
J. Clean. Prod. 174 (2018) 591–604.
[5] P. Priyadharshini, K. Ramamurthy, R.G. Robinson, Sustainable reuse of excavation soil in cementitious composites, J. Clean. Prod. 176 (2018) 999–1011.
[6] C.J. Engelsen, G. Wibetoe, H.A. van der Sloot, et al., Field site leaching from recycled concrete aggregates applied as sub-base material in road construction, Sci.
Total Environ. 427 (2012) 86–97.
[7] C. Shi, Y. Li, J. Zhang, et al., Performance enhancement of recycled concrete aggregate–a review, J. Clean. Prod. 112 (2016) 466–472.

11
L. Wei et al. Journal of Building Engineering 98 (2024) 111269

[8] S. Siddique, S. Shrivastava, S. Chaudhary, Influence of ceramic waste on the fresh properties and compressive strength of concrete, Eur. J. Environ. Civ. Eng. 23
(2) (2019) 212–225.
[9] U. Avila-López, J.M. Almanza-Robles, J.I. Escalante-García, Investigation of novel waste glass and limestone binders using statistical methods, Construct. Build.
Mater. 82 (2015) 296–303.
[10] R. Siddique, Waste Materials and By-Products in concrete[M], Springer Science & Business Media, 2007.
[11] K. Bisht, P.V. Ramana, Sustainable production of concrete containing discarded beverage glass as fine aggregate, Construct. Build. Mater. 177 (2018) 116–124.
[12] Q. Su, J. Xu, Mechanical properties of concrete containing glass sand and rice husk ash, Construct. Build. Mater. 393 (2023) 132053.
[13] M. Małek, W. Łasica, M. Jackowski, et al., Effect of waste glass addition as a replacement for fine aggregate on properties of concrete, Materials 13 (14) (2020)
3189.
[14] K. Bisht, P.V. Ramana, Sustainable production of concrete containing discarded beverage glass as fine aggregate, Construct. Build. Mater. 177 (2018) 116–124.
[15] I.S. Kim, S.Y. Choi, E.I. Yang, Evaluation of durability of concrete substituted heavyweight waste glass as fine aggregate, Construct. Build. Mater. 184 (2018)
269–277.
[16] ISO 679, Cement- Test Methods- Determination of Strength, International Standard, Switzerland, 2009.
[17] S. Huaishuai, Z. Junhao, Y. Guotao, Study on the bond behavior of corroded steel bars embedded in concrete under the coupled effect of reciprocating loads and
chloride ion erosion, Construct. Build. Mater. 305 (2021).
[18] R. Ke, L. Wang, L. Zheng, et al., Deterioration of mechanical properties at the anchor-concrete interface considering the effect of two-phase corrosion, Construct.
Build. Mater. 370 (2023) 130664.
[19] J. Hesong, L. Zhenlin, Z. Weizhuo, et al., Iodide and chloride ions diffusivity, pore characterization and microstructures of concrete incorporating ground
granulated blast furnace slag, J. Mater. Res. Technol. 16 (2022) 302–321.
[20] Y. Lee, T.Y. Joo, T. Lee, et al., Mechanical properties of constitutive parameters in steel–concrete interface, Eng. Struct. 33 (4) (2011) 1277–1290.
[21] Z.Z. De, X.S. Yun, L. Zhe, et al., Stress analysis of full-length grouted bolt under shear deformation of anchor interface, J. Mt. Sci. 19 (11) (2022) 3286–3301.

12