Abstract
The grouted sleeve connector is one of the most widely used connections in precast concrete shear wall structures. Due to the technology limit of the assembly workers, the sleeve grouting often has defects, which could significantly weaken their practical performance. Only considering the effect of defects on the bearing capacity of the sleeve connector cannot simulate the effect of defects in actual engineering. To this end, an experiment was carried out to investigate the seismic performance of precast concrete shear walls with sleeve grouting defects. Four full-scale precast concrete shear walls with different defect levels of insufficient grouting were designed and tested under cyclic loads. The development of concrete cracking, failure mode, hysteresis features, load-bearing capacity, stiffness degradation, structural ductility and energy-dissipation capacity are analyzed to reveal the influence of the sleeve grouting defects on precast concrete shear walls. The test results indicated that with the increase of the defect level, the opening gap between the bottom of the wall and the top of the foundation beam widened, and the failure mode changed from tensile fracture of reinforcing steel bar to anchorage failure of the tensile reinforcement on the defective side. Meanwhile, as the level of grouting defects increased, the bearing capacity decreased, and the specimen failed prematurely. The effect of the grouting defects on cracking load was minor, while they had a significant influence on the yield bearing capacity, peak bearing capacity and ultimate bearing capacity. The experimental results also indicated that the grouting defects caused the weakening of the stiffness, ductility and energy dissipation of the precast concrete shear walls. The presented work could be used for defect prevention and evaluation of precast concrete shear walls with sleeve grouting defects.
Keywords
Introduction
Due to the benefits of high construction efficiency, high degree of industrialization, green environmental protection and so on, precast shear wall structures have been extensively investigated and put into engineering appliance for decades (Buddika and Wijeyewickrema, 2018; Kurama et al., 1999; Peng et al., 2016; Smith et al., 2013; Xia et al., 2021). The vertical connection forms of prefabricated shear wall structures could significantly influence the overall performance of the structures. Consequently, the researchers have paid more attention to the connection of the precast shear wall structures (Lu et al., 2016; Peng et al., 2020; Su et al., 2021; Wang et al., 2022; Xiao et al., 2021). More recently, due to several advantages, including controllable tolerance, accelerated construction and environmental friendliness, the grouted sleeve connector has attracted the interest of researchers and engineers (Lu et al., 2018). The grouted sleeve connector is realized by inserting two reinforcing bars into a grouted sleeve connector and injecting grouting material into the connector. When the grouting material hardens, reinforcing bars and sleeve connector are firmly combined (Guo et al., 2022).
The grouted sleeve connectors are critical to ensuring that the performance of precast concrete structures is comparable to that of cast-in-situ structures. Since the grouted sleeve connector was invented by Yee in 1967 (1968), the mechanical properties of the grouting sleeve connector have been investigated for decades. Lamport et al. (1991) tested 18 grouted pile-to-sleeve connections to investigate the influence of combined loading, relative shear-key location, pile-sleeve eccentricity, and grout strength on their performance. The results indicated that combined loading had no detrimental effect on the ultimate capacity of a grouted connection, and that increasing shear-key height or decreasing shear-key spacing might not lead to higher ultimate strength of the connector. Hayashi et al. (1993) investigated the bond stress-slip characteristics of reinforcing bars in grouted sleeve connectors and analyzed the relationship between the local maximum bond stress and the slip of reinforcing bar. Ling et al. (2012) investigated the behavior of a grouted pipe splice under incremental tensile load and found that the geometrical configurations of the sleeve had a significant influence on bond performance. The previous researches showed that the grouted sleeve connector could provide excellent connection performance with careful design and fabrication.
In order to further investigate the influence of grouting sleeve connection on the performance of overall structures, some scholars have studied the seismic performance of precast shear walls with grouted sleeve connections. Peng et al. (2016) studied the seismic performance of precast concrete shear walls with a mortar-sleeve connection for longitudinal steel bars, discovering that the mortar-sleeve splicing effectively transferred stresses to the vertical steel bars. Li et al. (2019) investigated the seismic performance of a T-shaped partly precast reinforced concrete shear wall with grouting sleeves and found that the T-shaped partly precast reinforced concrete shear wall could be used in building structures to play the same role as cast in situ components under effective and reliable design. Wu et al. (2020) also investigated the seismic performance of precast short-leg shear walls with grouting sleeve connections. The test results revealed that the reinforcement bars connected by grouting sleeves were reliable before yielding, but the bonding strength between the grouting material and the connecting reinforcement bar in the sleeve decreased after yielding. According the previous researches, the precast concrete shear wall with the sleeve grouted connection has outstanding seismic performances under effective and reliable design.
Due to the backward technology of the assembly workers, the sleeve grouting often has defects, which could significantly weaken their practical performance. Several causes could lead to grouting defects in practical engineering, including insufficient anchor length of reinforcement, position deviation of reinforcement, insufficient grouting, et al. To evaluate the bond behavior of insufficiently grouted sleeve connection, Xu et al. (2018) proposed experiments to investigate the influences of defect level of insufficient grouting configuration (uniform, longitudinal, radial and inclined) on the bond failure mode and bond stress-slip curves. The bond strength and slip were found to decrease as the defect level increased, and empirical expressions for both the normalized bond strength and slip were presented. Zheng et al. (2020) investigated the mechanical performance of insufficiently grouted sleeve connections with predesigned vertical grouting defects and found that the failure mode of defective specimens might shift from tensile fracture of reinforcing steel bar to interfacial bond-slip failure of rebar as the anchorage length of reinforcing steel bar decreased. Guo et al. (2022) presented a systemic experimental investigation on the connection performance of fully-grouted sleeve connectors with various grouting defects, and found that their bearing capacity and deformation capacity decreased as total effective anchor lengths decreased.
Only considering the effect of defects on the bearing capacity of the sleeve connector cannot simulate the effect of defects in actual engineering. To this end, an experiment was carried out to investigate the impact of sleeve grouting defects on the seismic behavior of precast concrete shear walls. Four full-scale precast concrete shear walls with different defect levels of insufficient grouting were designed and tested under cyclic loads. For the specimen PSW 1 without sleeve grouting defect, the length of the reserved reinforcement of the foundation beam was 132 mm (8d + 20 mm). For the specimens PSW 2, PSW 3 and PSW 4, the reserved length of reinforcement at the defects was 76 mm (4d + 20 mm), and the number of reinforcements at the defects was different. The development of concrete cracking, failure mode, hysteresis features, load-bearing capacity, stiffness degradation, structural ductility and energy-dissipation capacity are analyzed to reveal the influence of sleeve grouting defects on precast concrete shear walls. The research presented in this paper could provide a technical basis for evaluation of the seismic behavior of precast concrete shear walls with sleeve grouting defects.
Experimental program
Specimen design and fabrication
The test specimens’ wall panels were 1700 mm long, 200 mm wide and 2800 mm tall. They were supported by foundation beams that were on 2700 mm long, 900 mm wide and 700 mm tall. The 1700 mm long, 260 mm wide and 260 mm tall beam was integrally cast with the wall panel at the top to permit the introduction of lateral load, as shown in Figure 1. The wall panels had the same reinforcement details for all specimens, as shown in Figure 2. The wall panels were equipped with boundary elements consisting of six 14 mm bars confined with 8 mm stirrups at 200 mm. The boundary reinforcement confinement spacing was reduced to 100 mm in the lower 550 mm of the wall panels. Ten 10 mm bars were distributed in the middle of the wall panels confined with 100 mm stirrups at 200 mm, and the confinement spacing was reduced to 100 mm in the lower 550 mm of the wall panels. Eight additional 14 mm with a length of 830 mm (700 mm + 130 mm) were distributed on both sides of the middle wall. The 20 14 mm vertical bars were grouted into sleeve connectors for attachment to the foundation beam. Configuration of the commercial fully-grouted sleeve connectors is shown in Figure 3, and their dimensions are listed in Table 1. L
1
denotes the anchor length of the reinforcing bar assembled on construction site (i.e., the bottom reinforcing bar), and L
2
denotes the anchor length of the reinforcing bar embedded in prefabricated factory (i.e., the top reinforcing bar). Dimensions of the specimens. Reinforcement details of wall panels: (a) Layout of steel reinforcements; (b) Section 1; (c) Section 2; (d) Section 3; (e) Section 4. Configuration of the fully-grouted sleeve connector. Detailed dimensions of the fully-grouted sleeve connector (Unit: mm).


According to the researches conducted by Xu et al. (2018) and Guo et al. (2022), reducing the total effective anchor lengths of the bottom reinforcing bars could significantly deteriorate anchor properties and change the failure mode from fracture failure to anchor failure. The previous numerical simulation showed that vertical reinforcement in the boundary elements of the shear wall panel had a significant influence on the bearing capacity of the precast shear wall, while vertical reinforcement in the middle of the wall panel had little influence. Based on the findings, four full-scale precast concrete shear walls with different sleeve grouting defects were designed, namely, PSW 1, PSW 2, PSW 3 and PSW 4. Meanwhile, this test provided the defects in the bars of the boundary elements, and only on one side of the boundary elements. For the specimen PSW 1 without sleeve grouting defect, the length of the reserved reinforcement of the foundation beam was 132 mm (8d + 20 mm), where 20 mm was the thickness of the bed-mortar layer and 8d was the anchorage length of the reinforcement extending into the sleeve (d was the diameter of the reinforcement). For the specimens PSW 2, PSW 3 and PSW 4, the reserved length of reinforcement at the defects was 76 mm, that is, the anchorage length of reinforcement extending into the sleeve was 4d. Figure 4 shows the defect details of each specimen. Defect details of each specimen.
The fabrication process of the specimens was as follows: (1) Preparing the steel cages of foundation beams and wall panels, as shown in Figure 5(a) and (b); (2) Installing the connecting pipes to the grouting hole and vent hole of the sleeve connectors, as shown in Figure 5(c); (3) Casting foundation beams and wall panels, as shown in Figure 5(d) and (e); (4) Casting the bed-mortar layer, as shown in Figure 5(f); (5) Assembling the wall panel to the foundation beam, as shown in Figure 5(g); (6) Casting grouting material to the sleeve connectors, as shown in Figure 5(h). Fabrication process: (a) Steel cages of foundation beams; (b) Steel cages of wall panels; (c) Installing the connecting pipes; (d) Prefabricating foundation beams; (e) Prefabricating wall panels; (f) Casting bed-mortar layer; (g) Attaching the wall panel to the foundation beam; (h) Casting grouting material.
Material properties
Mechanical properties of the steel bars (Unit: mm or MPa).
Test setup and loading process
The cyclic load at the top of the specimens was applied by using a horizontal actuator located on the south side of the specimens, as shown in Figure 6. The actuator was set to the positive direction when pushed, while the negative direction when pulled. A hydraulic jack was used to apply a constant axial force (1000kN) to the top of precast concrete shear walls, and the foundation beam was fixed to the laboratory floor. The axial compression ratio was 0.2 for all specimens. The loading protocol adopted displacement-controlled pattern, as shown in Figure 7, with 15 loading levels and the corresponding displacement amplitude of 2 mm, 4 mm, 6 mm, 8 mm, 10 mm, 15 mm, 20 mm, 25 mm, 30 mm, 35 mm, 40 mm, 45 mm, 50 mm, 55 mm, and 60 mm, respectively. The first four levels were cycled twice per level, and the last 11 levels were cycled three times per level. The loading direction of the specimens is shown in Figure 8. The test was terminated when the applied load fell below 85% of the peak load or the hysteresis loop became unstable. Test setup: (a) Loading system and instrumentation; (b) Specimen under test. Displacement-controlled loading protocol. Schematic diagram of loading direction of the specimens: (a) PSW 1; (b) PSW 2; (c) PSW 3; (d) PSW 4.


Instrumentation
The transducer layout is shown in Figure 6. Two displacement transducers (D-1 and D-2) were placed at the centerline of the loading beam to measure the top horizontal displacement. A horizontal displacement transducer (D-3) was placed on the foundation beam to measure the translation, and two vertical displacement transducers (D-5 and D-6) were placed on the foundation beam to measure the rotation.
Test progression and failure modes
Specimen PSW 1
Figure 9 shows the cracking and failure mode of the specimen PSW 1. When the horizontal force reached 484kN (positive direction), two horizontal cracks appeared on the south side of the wall. The first horizontal crack was about 250 mm from the bottom of the wall and about 480 mm long. The second horizontal crack was about 450 mm from the bottom of the wall and about 250 mm long. When the horizontal force reached −381kN (negative direction), a 370 mm long horizontal crack appeared on the north side of the wall, about 250 mm from the bottom of the wall. In the ninth level of cyclic loading, a horizontal seam appeared through the bed-mortar layer of the wall, and the precast wall and foundation beam began to separate and open. In the 10th stage of cyclic loading, the cracks on the north and south sides of the wall continued to extend, with the cracks on the lower part of the south wall extending obliquely, and the bottom of the wall and the foundation beam were obviously open, with an opening width of 5 mm. During the 14th and 15th levels of cyclic loading, the concrete protective layer at the bottom of the north wall spalled. Due to the maximum tensile force limit of the horizontal actuator, the compressive damage degree of the concrete on the south side was lower than that on the north side. During the 16th level of cyclic loading, the horizontal force fell below 85% of the peak bearing capacity, and there was a sound of reinforcing steel bar tensile fracture. The fracture point was located at the intersection of the wall bottom and the foundation beam. Furthermore, the sudden tensile fracture of reinforcing steel bar caused out-of-plane deviation on the south side of the wall. The failure mode of the specimen PSW 1 was compression bending failure. The tensile reinforcement was broken, and the compressed concrete was crushed and spalled. The crack distribution was relatively uniform. Due to the maximum tensile force limit of the horizontal actuator, the cracks on the south and north sides were asymmetric, and the tensile damage on the south side was more severe. Cracking and failure mode of specimen PSW 1: (a) Cracking; (b) Failure mode.
Specimen PSW 2
Figure 10 shows the cracking and failure mode of the specimen PSW 2. When the horizontal force reached 306kN (positive direction), the first horizontal crack appeared on the south side of the wall. The first horizontal crack was about 250 mm from the bottom of the wall and about 300 mm long. When the horizontal force reached −403kN (negative direction), a 400 mm long horizontal crack appeared on the north side of the wall, about 250 mm from the bottom of the wall. In the eighth level of cyclic loading, a horizontal seam appeared through the bed-mortar layer of the wall, and the precast wall and foundation beam began to separate and open. During the 15th level of cyclic loading, the reinforcement in the sleeve connector on the defective side was pulled out, and the concrete on both sides of the wall was crushed and spalled. During the 16th level of cyclic loading, the horizontal force fell below 85% of the peak bearing capacity. The failure mode of the specimen PSW 2 was compression bending failure. Cracking and failure mode of specimen PSW 2: (a) Cracking; (b) Failure mode.
Specimen PSW 3
Figure 11 shows the cracking and failure mode of the specimen PSW 3. When the horizontal force reached 327kN (positive direction), the first horizontal crack appeared on the south side of the wall. The first horizontal crack was about 250 mm from the bottom of the wall and about 450 mm long. When the horizontal force reached −356kN (negative direction), a 500 mm long horizontal crack appeared on the north side of the wall, about 250 mm from the bottom of the wall. In the eighth level of cyclic loading, a horizontal seam appeared through the bed-mortar layer of the wall, and the precast wall and foundation beam began to separate and open. During the 15th level of cyclic loading, the reinforcement in the sleeve connector on the defective side was pulled out, and the concrete on both sides of the wall was crushed and spalled. During the 16th level of cyclic loading, the horizontal force fell below 85% of the peak bearing capacity. The failure mode of the specimen PSW 3 was compression bending failure. Cracking and failure mode of specimen PSW 3: (a) Cracking; (b) Failure mode.
Specimen PSW 4
Figure 12 shows the cracking and failure mode of the specimen PSW 4. When the horizontal force reached 340kN (positive direction), the first horizontal crack appeared on the south side of the wall. The first horizontal crack was about 250 mm from the bottom of the wall and about 250 mm long. When the horizontal force reached −315kN (negative direction), a 150 mm horizontal crack appeared on the north side of the wall, and about 250 mm from the bottom of the wall. In the eighth level of cyclic loading, a horizontal seam appeared through the bed-mortar layer of the wall, and the precast wall and foundation beam began to separate and open. During the 15th level of cyclic loading, the reinforcement in the sleeve connector on the defective side was pulled out, and the concrete on both sides of the wall was crushed and spalled. Meanwhile, the horizontal force fell below 85% of the peak bearing capacity. The failure mode of the specimen PSW 4 was compression bending failure. Cracking and failure mode of specimen PSW 4: (a) Cracking; (b) Failure mode.
During the loading process, the phenomenon of the opening between the bottom of the wall panel and the top of the foundation beam occurred. Affected by the sleeve connectors, the first crack of the specimens appeared at the height of the top surface of the sleeve connectors, and finally formed horizontal cracks penetrating the specimens. Within the height range of the sleeve connectors, there was no crack on the wall panels generated from both sides of the boundary elements. In conclusion, as the defect level increased, the opening gap between the bottom of the wall and the top of the foundation beam became larger, and the failure mode changed from the tensile fracture of reinforcing steel bar to the anchorage failure of the tensile reinforcement on the defective side.
Results analysis and discussion
Hysteresis curves and envelope curves
Figure 13 shows the hysteresis curves and envelope curves of the four specimens. Although the push capacity of the actuator is 1000kN, its maximum tensile force could only reach about 650kN and increasing the tensile force could cause the actuator to be damaged. Therefore, when the negative target displacement was large, the manual control pattern was adopted. Before yielding, the hysteresis loops were narrow and had a small enclosed area, indicating that the specimens were in elastic working stage. After yielding, as the horizontal displacement increased, the specimens’ cracks extended and their lateral stiffness degraded. Meanwhile, the hysteresis loops became plump, and energy dissipation increased. For comparison, Figure 14 shows the envelope curves of all specimens. When the horizontal displacement was small, the grouting defect has little effect on the mechanical performance of the specimens. as the defect level increased, the bearing capacity decreased, and the specimen failed prematurely. Meanwhile, the rate of decline in the bearing capacity slowed down. Hysteresis curves and envelope curves: (a) PSW 1; (b) PSW 2; (c) PSW 3; (d) PSW 4. Envelope curves of all specimens.

Bearing capacity analysis
The peak displacement and corresponding yield displacement for Figure 15 can be calculated using the equivalent energy method. For the origin displacement-load curve OABCG, replace the continuous curve OABCG with the broken line ODC. When the two shade areas of OAB and BDCE are identical, the point E is regarded as the yielding point of the specimen (Park, 1989; Yang et al., 2020). Figure 16 shows the characteristic points of tested specimens and the relationship between the values of characteristic points and the total effective anchor lengths (L) of the bottom reinforcing bars in the boundary elements (defect side). The specimen PSW 1 had a higher positive cracking load. This was because the cracking load was taken as the load value when the first crack appeared. When the positive load reached 484.1kN, two cracks appeared in the specimen PSW 1. Therefore, the load value of the first crack in the specimen PSW 1 was between 283.6kN and 484.1kN. From the test results in Figure 16, the effect of the grouting defects on cracking load was small. Taking the yield load and peak load of the specimen PSW 1 as the reference standard values, the yield load of the specimen PSW 2 was 87% of the specimen PSW 1, and the peak load was 89% of the specimen PSW 1. The yield load of the specimen PSW 3 was 81% of the specimen PSW 1, and the peak load was 85% of the specimen PSW 1. The yield load of the specimen PSW 4 was 77% of the specimen PSW 1, and the peak load was 83% of the specimen PSW 1. Affected by the grouting defects, the ultimate loads of the defective specimens PSW 2, PSW 3 and PSW4 were slightly lower than that of the specimen PSW 1, which were 95%, 94% and 90% of the specimen PSW 1, respectively. As the total effective anchor lengths (L) decreased, the linear decrease in the load characteristic values was obvious, especially for the yield load, peak load and ultimate load. Meanwhile, the decrease rate of yield load, ultimate load and residual strength decreased gradually. The downtrends about the displacement values were not obvious. Therefore, the design defects had a significant influence on the yield bearing capacity, peak bearing capacity and ultimate bearing capacity of the precast concrete shear walls. Determination of yield point. Relationship between the values of characteristic points and the total effective anchor lengths (L): (a) Crack point; (b) Yield point; (c) Peak point; (d) Ultimate point.

Stiffness degradation
The stiffness degradation is an important indicator for evaluating structural damage. To analyze the stiffness degradation pattern, the secant stiffness in each load level is defined as follows:
According to equation (1), the secant stiffness-drift relationships of the specimens were obtained, as shown in Figure 17. The variation trend of secant stiffness was essentially the same for all specimens. When the displacement was small, due to the virtual displacement caused by the installation error of the specimens, the initial secant stiffness of the specimen PSW1 was lower than that of the specimens with grouting defects. In the negative direction, as displacement increased, the secant stiffness of all specimens remained essentially constant. In the positive direction, as displacement increased, the stiffness of the specimen PSW1 was slightly larger than that of the specimens with grouting defects. Degradation curves for secant stiffness.
Displacement ductility
Ductility coefficients of specimens in the positive direction (Unit: mm).
Energy dissipation
The energy-dissipation capacity is a critical factor to elaborate the structural aseismic capacity. In this study, the equivalent viscous damping coefficient (ζeq) is adopted to calculate energy dissipation, as shown in . Diagram of equivalent viscous damping coefficient.

Equivalent viscous damping coefficient in the ultimate cycle stage.
Conclusion
This study reported four reversed cyclic load tests on full-scale precast concrete shear walls with different levels of insufficient grouting. The development of concrete cracking, failure mode, hysteresis features, load-bearing capacity, stiffness degradation, structural ductility and energy-dissipation capacity are analyzed to reveal the influence of the sleeve grouting defect on precast concrete shear walls. Based on the test results, the following conclusions can be drawn: (1) As the defect level increased, the opening gap between the bottom of the wall panel and the top of the foundation beam became larger, and the failure mode changed from the tensile fracture of reinforcing steel bar to the anchorage failure of the tensile reinforcement on the defective side. (2) As the level of grouting defects increased, the bearing capacity decreased, and the specimen failed prematurely. Meanwhile, the bearing capacity of the specimen declined at a slower rate. (3) The effect of the grouting defects on cracking load was minor, while had a great influence on the yield bearing capacity, peak bearing capacity and ultimate bearing capacity of precast concrete shear walls. (4) In general, the grouting defects weakened the stiffness, ductility and energy dissipation of precast concrete shear walls.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China under grant No. 52125802.
