Abstract
To study the cyclic behavior of partially precast steel fiber high-strength concrete columns with high-strength steel bars, four full-sized square column specimens were fabricated and tested under constant axial load and horizontal cyclic load. The effects of the strength of precast concrete shell and the diameter of cast-in-place column core were analyzed in detail. The results show that partially precast steel fiber high-strength concrete columns have good seismic performance and deformation ability. Compared to the concrete column with lower strength of precast concrete shell, the concrete column with higher strength of precast concrete shell showed higher bearing capacity and energy dissipation capacity while lower ductility. Moreover, with the increase of the diameter of cast-in-place column core, the bearing capacity and the deformation ability of the specimen decreased. Finally, based on the experimental research and theoretical analysis, a calculation model for predicting the maximum bearing capacity was proposed, and the results obtained from the formulas were in good agreement with those from the experiments.
Keywords
Introduction
Construction is one of the most energy intensive industries (Lai et al., 2016), and construction waste accounts for 25%–33% of all waste generated in the EU. The problem that buildings consume more than 35% of primary energy around the world (UNE and IEA, 2017), can be alleviated by prefabrication construction effectively. Prefabrication refers to the process that building components in the factory, transporting complete or half components to the construction site, and assembling the components to build the construction. Compared with the traditional construction technology, prefabricated construction technology has the advantages that protecting the environment, reducing on-site construction difficulty, ensuring construction quality, reducing project cost, and facilitating construction management (Jaillon and Poon, 2009; Jeong et al., 2017; Tam et al., 2015). The research shows that compared with on-site construction (Hong et al., 2016; Zhu et al., 2018), prefabricated buildings can reduce resource consumption, health damage and ecosystem damage by 35.82%, 6.61%, and 3.47%, respectively.
Prefabricated components have been well developed and widely used in a large number of developed countries such as Germany, the Netherlands, Denmark, and Sweden, the prefabrication coverage of which are severally 31%, 40%, 43%, and 80% (Cheng et al., 2017; Mao et al., 2013). With the acceleration of China’s urbanization process and the increase of population (Fan et al., 2017), China has formulated a series of policies, such as “Several Opinions on Further Strengthening the Management of Urban Planning Construction” and “13th Five-Year Prefabricated Construction Action Plan” to promote the application in the construction industry. Scholars have carried out a series of research on prefabricated components. Zhang et al. (2019a) carried out seismic performance test on the joint of prefabricated steel plate shear wall and cover plate. Li et al. (2020a) studied the seismic performance of prefabricated concrete beam-column joints, with better energy dissipation. Hu (Hu et al., 2020) has carried out the low cycle reciprocating test on three 1/2 scale prefabricated structures, and found out the prefabricated members can achieve better seismic performance. At the same time, relevant research pointed out that the transportation and installation cost of the whole prefabricated component is high, and the quality of component connection is not easy to guarantee (Xue et al., 2019).
As an alternative method to reduce the deadweight of precast members, partially precast members were studied. Hong (Hong et al., 2008) put forward a structure named “a partially prefabricated reinforced concrete composite beam” and carried out a systematic study. Yang (Yang et al., 2017, 2018, 2019) conducted low-cycle reciprocating tests on partially prefabricated concrete columns. Du (Chuang et al., 2020) studied seismic behavior of concrete-filled precast tubular column, and the nominal cubic compressive strength (GB50010, 2010) of the concrete is 30–50 MPa, while the steel strength grade is HRB400 (i.e. yield strength of 400 MPa and higher). The results showed that the column has excellent seismic properties. However, few research has been conducted on precast high-strength concrete columns with high-strength steel bars. The use of high-strength concrete can reduce the section size of components and increase the use area of buildings (Teng and Chandra, 2016). Using high-strength steel bars can reduce the number of steel bars and avoid reinforcement congestion, so as to achieve optimal design and reduce cost. Some studies showed that the bearing capacity of the specimen can be effectively improved and the residual deformation can be reduced by using high-strength reinforcement (Li and Aoude, 2020b). Compared with the specimens with ordinary steel bars, the columns with high-strength steel bars can achieve similar bearing capacity, ductility, and energy dissipation capacity while reducing the number of steel bars (Rautenberg et al., 2012). It has important social and economic value to study partially precast high strength concrete columns with high strength steel bars.
In this study, a new type of partially precast concrete column is proposed, which is composed of precast concrete shell and cast-in-place column core as shown in Figure 1. Yang (Yang et al., 2018) pointed out the concept and installation process of partially precast structural system. As steel fiber can improve the tensile strength of concrete and delay the cracking of concrete (Parra-Montesinos, 2015), it is used in precast concrete shell tentatively. The previous experiment carried out by the author (Zhang et al., 2019c) showed that compared with the partially precast columns with ordinary steel bars, the bearing capacity and energy dissipation capacity of the specimens with high-strength steel bars were increased. Based on the above research, this paper will study the effects of the strength of precast concrete shell and the diameter of cast-in-place column core on partially precast concrete columns. To the knowledge of authors, these two parameters have not been studied. The main objectives of this research are (1) to study the failure mode, hysteretic performance, deformation capacity, stiffness, and energy dissipation capacity of the partially precast steel fiber high-strength concrete columns with high-strength steel bars; (2) to analyze the effects of the strength of precast concrete shell and the diameter of cast-in-place column core on the cyclic behavior; (3) to propose a simplified calculation model for predicting the maximum bearing capacity on the basis of experimental observation and test results.

Schematic diagram of partially precast concrete components: (a) cross section and (b) stereogram.
Experimental program
Design of test specimens
Four partially precast HRB600 reinforced steel fiber high-strength concrete columns were designed and tested. In order to fully consider the engineering practice, the size of the specimen is full-scale and the design axial compression ratio is 0.65. Partially precast concrete column is composed of precast concrete shell and cast-in-place column core. The main parameters of the specimens were summarized in Table 1. The variables were the strength of precast concrete shell and the diameter of cast-in-place column core. Compared with specimen PC-2, the strength of precast concrete shell of specimen PC-1 is C60, and the diameters of cast-in-place column core of specimen PC-3 and PC-4 are 350 and 450 mm, respectively.
Basic parameters of specimens.
Figure 2 shows the cross-sections and steel reinforcement details of the specimens. A total of 24 D16 (16 mm diameter) steel bars were used as longitudinal bars in the columns. D10 (10 mm diameter) stirrups were placed spacing of 100 mm. All steel bars were built into the precast concrete shell. First, the precast concrete shell was processed in the factory with steel formwork and the reinforcement at the top and bottom of the column was reserved for assembling and connecting with the loading beam and the foundation respectively. Then the bottom of precast concrete shell was spliced with the foundation. Next, the top of precast concrete shell was assembled with the loading beam, including assembling the reinforcement of the loading beam and pouring concrete. When pouring the concrete of the loading beam, the concrete column core was poured together. When pouring the cast-in-place column core, the precast concrete shell and foundation acted as the formwork.

Dimension and reinforcements of specimens.
Material properties
Concrete
Three types of concrete with strength grade of C60 and C80 were adopted in this study. The precast shell concrete was C60 or C80 steel fiber high-strength concrete. The cast-in-place column core concrete was C80 ordinary high-strength concrete. The loading beam and foundation of each specimen used the same type of concrete as the cast-in-place column core. Due to the large axial and horizontal forces during loading process, 6 mm thick steel plate is covered on the top of the column and the outside of the foundation to ensure sufficient stiffness. Common Portland cement (P.O.52.5R) was chosen to ensure strength. Gravel and natural river sands with particle size ranging from 5 to 20 mm were used as coarse aggregate and fine aggregate, respectively. About 1.5% volume ratio of steel fiber used in concrete was selected in order to ensure good performance of PC in this test (Zhao et al., 2018). As shown in Figure 3, the equivalent diameter, average length and tensile strength of the hook-end steel fiber were 0.55, 35, and 1350 MPa, respectively. The mixing proportion and the mechanical properties of three types of concrete measured according to relevant standards (CECS38, 2004; GB50010, 2010) are shown in Table 2.

Illustration of the hook-end steel fibers.
Mix proportion and mechanical properties of concrete.
fcu is the measured cubic compressive strength; fc is the prismatic compressive strength of concrete; ft is the tensile strength of concrete; Ec is the elastic modulus of concrete.
Steel
Two kinds of HRB600 steel bars with different diameters were used in the test. The mechanical properties of steel bars were determined according to the national standard (GB/T228.1, 2010), as summarized in Table 3.
Mechanical properties of reinforcement.
Test setup and measurements
Specimens were tested under horizontal cyclic load and constant axial load, and the test setup is shown in Figure 4. The experiment was carried out in the multifunctional electro-hydraulic servo loading test system as shown in Figure 5. It can be seen that the vertical compressive loading was applied by a 40000 kN hydraulic jack, and the cyclic lateral loading was applied at the loading point by the 2000 kN electro-hydraulic servo machine tool. At the same time, Figure 5 shows the arrangement of linear variable differential transformers (LVDTs) on a specimen. Three horizontal LVDTs were applied to measure the lateral displacements at various heights (i.e. D1, D2, and D3). D1 and D2 were used to measure the horizontal displacement of the top of the column and the foundation, respectively. D3 was used to measure the relative displacement between the foundation and the skateboard base. As shown in Figure 6, strain gauges were set on the longitudinal bars and stirrups, which were 50 and 150 mm away from the surface of the foundation.

Experimental test setup.

Schematic diagram of test setup.

Arrangement of strain gauges on steel bars.
The test axial compression ratio was 0.25, which is calculated as stipulated by the Chinese code. The axial force of PC-1 was 5400 kN and that of others was 7000 kN. As shown in Figure 7, the cyclic lateral load was controlled by the drift ratio θ, θ=Δ/H, where Δ is the horizontal displacement measured by a linear variable differential transformer (D2), and H is the height from the center of the spherical hinge to the surface of the foundation, which was 2150 mm. A preload was applied with a 0.125% drift ratio to confirm the normal operation of the loading device before the formal loading. Then the increase of drift ratio can be divided into three stages: first, increasing with a increment of 0.25% until the drift ratio reaches 1%; then raising increment to 0.5% until reached 4% drift ratio; last, increasing the increment to 1% until the end of the test has been terminated. Each step is loaded twice until a drift rate of 2% is reached, and then each step is loaded once. The loading process was terminated when the specimen lost its vertical bearing capacity or the severe damage occurred.

Loading program.
Test results and analysis
Failure modes
Figure 8 shows the damage patterns of all specimens measured at 2% drift ratios and Figure 9 shows the final damage patterns. In general, all specimens were damaged by bending. At the beginning of loading, there was little damage to the specimen. When the drift ratio reached 0.5% to 1%, small horizontal cracks appeared within 100–400 mm from the bottom of column. With the increase of loading displacement, the number, width, and length of cracks gradually increased. The crack distribution range gradually increased along the column height and concrete crushed in compression area. When specimen approaches failure mode, the spalling and crushing of concrete and buckling of some longitudinal bars occurred in different degrees, thus specimens would lose the bearing capacity. Because the steel fiber has the function of crack resistance, the crack development of the specimen is relatively uniform and only a small piece of concrete flaked off at the bottom of the specimen. During the experiment, the deformation of the column is coordinated and there is no vertical bond crack. The precast concrete shell and cast-in-place column core achieve good cooperative performance.

Cracks distribution of specimens with 2% drift angle: (a) PC-1, (b) PC-2, (c) PC-3, and (d) PC-4.

Failure modes of specimens: (a) PC-1, (b) PC-2, (c) PC-3, and (d) PC-4.
Hysteretic behavior
The hysteresis curve of all specimens are shown in Figure 10, where F is the measured horizontal load at the loading point. Δ and θ are the horizontal displacement and drift ratio at the loading point, respectively. It can be seen from Figure 10 that in the initial loading stages, the hysteresis curves behaved linear-elastically due to the negligible damage of materials. As the loading displacement increased, the area enclosed by hysteresis curves increased gradually, and the residual deformation increased.

Load-displacement hysteretic loops of specimens: (a) PC-1, (b) PC-2, (c) PC-3, and (d) PC-4.
The skeleton curves of all specimens are shown in Figure 11. Table 4 shows all recorded characteristic loads and the corresponding drift ratios, as the average value of load in positive and negative directions. The measured cracking load, Fcr, is defined as the load when initial cracking occurred. The yield load, Fy, is determined by the energy equivalent method (Park, 1989). The peak load, Fp, is defined as the max load during loading and the ultimate load, Fu, is defined as the load decreased to 85% of the peak load, Fp (or the corresponding load when specimen failure occurred). The ductility coefficient μ is defined as the ratio of ultimate drift ratio θu,t to yield drift ratio θy,t. The ultimate drift ratio of each specimen was between 1.84% and 3.35%, which was far higher than the elastic-plastic drift limit of 0.83% specified in national standard GB50011 (2010). The results indicate that all specimens have good deformation ability and the influence of different factors on the bearing capacity and deformation capacity of the specimen is discussed as follows.

Comparison of skeleton curves: (a) comparison between PC-1 and PC-2 and (b) comparison between PC-2, PC-3, and PC-4.
Experimental results of specimens at characteristic point.
Effect of the strength of precast concrete shell
The yield load and peak load of specimen PC-2 were 19.5% and 19.6% higher than those of specimen PC-1, respectively. With the rising of the strength of precast concrete shell, the concrete intended to perform a lower ductility. As we can see from Figure 11(a), the descending section of the skeleton curve dropped rapidly, and the ductility coefficient and ultimate drift ratio decreased by 7.3% and 17.6%, respectively.
Effect of the diameter of cast-in-place column core
With the increase of the diameter of cast in place column core, the bearing capacity and ductility of the specimen decreased gradually. Compared with PC-3, the peak bearing capacity of PC-2 and PC-4 was reduced by 2% and 4.6% respectively, and the ductility coefficients of PC-2 and PC-4 were reduced by 6% and 9.5% respectively. As shown in Figure 11(b), since the contact area of precast concrete shell and cast-in-place column core expanded as the diameter of cast-in-place column core increases, the column would show unsatisfied performance.
Strain analysis
The measured curve of strain displacement angle of longitudinal reinforcement is shown in Figure 12. Only strain data in the range of 4% displacement angle are given since the strain gauge is damaged under large deformation. It can be known from the Figure 12 that at the initial stage of loading, the strain of longitudinal reinforcement increases with the displacement angle until the reinforcement yields. As the displacement angle continues to increase, the steel bar began to develop plasticity and the strain increases rapidly. The results show that HRB600 steel bar can make full use of the tensile and compressive strength. Figure 13 shows the strain-displacement angle relationship of PC-2 and PC-4 stirrups. The stirrup will not yield until the specimen is destroyed, which indicates that the specimen had enough shear capacity.

Longitudinal bar strains: (a) PC-1, (b) PC-2, (c) PC-3, and (d) PC-4.

Stirrup bar strains: (a) PC-2 and (b) PC-4.
Stiffness degradation
The relationship between the measured scant stiffness K and the drift ratio θ is shown in Figure 14. In general, the stiffness degradation trend of each specimen was relatively close. With the concrete cracking and steel yielding, the stiffness degradation rate increases until the drift ratio reaches 2%, and then the stiffness degradation tended to be gentle. Due to the low strength of precast concrete shell, the stiffness of PC-1 is lower than that of other specimens during loading. It shows that the cooperative performance of the column is better when the strength of the precast concrete shell and the cast-in-place column core is both C80. In addition, the diameter of cast-in-place column core has little effect on the stiffness degradation of specimens.

Comparison of stiffness attenuation curves.
Energy dissipation capacity
The cumulative energy dissipation Ep and the equivalent viscous damping coefficient he were used to reflect the energy dissipation capacity of specimens, as shown in Figures 15 and 16. The cumulative energy consumption Ep refers to the total area enclosed by the hysteresis curves of all loading cycles in the first loading circle until the drift ratio reached a certain level. The equivalent viscous damping coefficient he is calculated according to Chopra (Chopra, 2007), which can reflect the integrity of the hysteresis curves. At the initial stage of loading, the specimen was in the elastic phase, and the loading and unloading curves were close, resulting in the meaningless of he. Figure 16 shows the relationship curves between he and θ, after reaching a drift ratio of 1%. It can be seen that the Ep and the he of each specimen tended to increase with drift ratios increasing. Compared with PC-1, the cumulative energy consumption and viscous damping coefficient of PC2 increased by 28.9% and 23.4%, respectively. The results show that the specimen with C80 strength of precast concrete shell has better energy consumption capacity. In addition, the curves of PC-2, PC-3, and PC-4 are close to each other, indicating that the diameter of cast in place column core has little effect on the energy consumption capacity.

Comparison of accumulated energy dissipation.

Equivalent viscous damping coefficient.
Seismic bending moment capacity
Superposition method
With reference to the calculation method of the bearing capacity of composite columns (Ji et al., 2014; Zhang et al., 2019b), the bending capacity of precast tubular high-strength concrete columns was calculated. It should be noted that, different from the composite columns in reference (Zhang et al., 2019b), the precast concrete shell and cast-in-place column core were not cast at the same time. In order to verify the applicability of the calculation method to the precast specimens, this influence factor is ignored. The bearing capacity of partially precast concrete columns is divided into two parts: core cast-in-place concrete columns and precast reinforced concrete columns, as calculated by equation (1). Firstly, the axial force of the precast concrete column was distributed according to equations (2) to (4), then the bending moments of the core cast-in-place concrete and the precast reinforced concrete under respective axial pressure were calculated, and the total bending moment of the section is obtained by adding the previous results, and the horizontal bearing capacity of the specimen was finally obtained.
where Nt is the test axial force, No and Ni are the axial bearing capacity of precast reinforced concrete columns and core cast-in-place concrete columns, respectively; λ is the distribution factor of axial force; fco, fci are the compressive strength of external concrete and core cast-in-place concrete; fy is the yield strength of longitudinal bars; Aco, As, Aci are the sectional area of external concrete, total longitudinal bars, and core cast-in-place concrete; Mt, Mo, Mi are the calculated ultimate moment of the section center provided by partially precast concrete column, precast reinforced concrete column, and core cast-in-place concrete column, respectively.
Capacity calculation of cast-in-place column core
The bearing capacity of the core cast-in-place concrete columns was calculated by the commonly-used sectional limit equilibrium theory. The calculation diagram of core cast-in-place concrete columns is shown in Figure 17. The Ni-Mi correlation curve was developed to established the corresponding bending moment under certain axial force.
where xn is the height of the compression zone, r is the radius of cast-in-place column core.

Calculation diagram of core cast-in-place concrete.
Capacity calculation of precast concrete shell
The precast concrete shell was converted into I-shaped section according to the principle of equivalent inertia moment of the section, as shown in the Figure 18. The basic assumptions adopted are as follows:
The plane section assumption was satisfied.
As shown in Figure 18(c), the compression zone of the concrete was simplified by using the equivalent rectangular stress block of average stress αcfco and compression zone depth βcxc, where xc denotes the actual depth of compression zone, and αc and βc are coefficients of the equivalent stress block defined in GB 50010 (2010). αc and βc respectively are 0.98 and 0.78 for C60 grade concrete and αc and βc respectively are 0.94 and 0.74 for C80 grade.
The longitudinal reinforcements at the tension zone were regarded as yield in tension, and the concrete in compression zone reached its ultimate compression strain, εcu. For nonfiber-reinforced specimens, εcu is 0.003 (GB 50010, 2010). For steel fiber-reinforced specimens, εcu is taken as 0.0035 to account for the effect of fiber on the ultimate compressive strain (Jin et al., 2018).
The tensile strength of steel fiber reinforced concrete was calculated according to CECS38 (2004), where the equivalent depth of tension zone xt is defined as xt=h − xc.
where εsi is the strain of the ith longitudinal rebar and xi is the distance between the ith longitudinal rebar and the extreme compression fiber. fsi and Asi represent the stress and the cross-sectional area of the ith longitudinal rebar, respectively. h is the height of the I-shaped cross section; b is the width of the cross section; and d is the diameter of cast-in-place column core.

Calculation diagram of precast reinforced concrete columns: (a) cross-section simplification, (b) strain distribution, and (c) stress distribution.
Capacity calculation results
Due to the large axial force and horizontal displacement of each specimen in the later stage of the test, the P-Δ effect has a great influence on the bending moment of the column bottom section which cannot be ignored. Therefore, equation (16) is used to calculate the horizontal bearing capacity, Fc.
where Δm is the horizontal deformation corresponding to the peak load of the specimen.
The calculated and measured results of horizontal bearing capacity of each specimen are listed in Table 5. It can be seen that the calculated result of each specimen is close to the test result, and the calculation error is not more than 10%. The formula is applicable to the calculation of the bearing capacity of partially precast concrete columns.
Comparison between experimental values and calculated results.
Conclusions
In this study, four partially precast steel fiber high-strength concrete columns with high-strength steel bars were investigated. The influences of the strength of precast concrete shell and the diameter of cast-in-place column core were analyzed and compared. The following conclusions can be drawn:
The partially precast steel fiber high-strength concrete columns exhibited good seismic performance and deformation ability. The precast concrete shell and cast-in-place column core can achieve good cooperative performance
Compared to the concrete column with lower strength of precast concrete shell, the concrete column with higher strength of precast concrete shell showed higher bearing capacity and energy dissipation capacity while lower ductility. In the design of partially precast steel fiber high-strength concrete columns, it is advised that C80 steel fiber reinforced concrete should be used in precast concrete shell when C80 concrete is used in cast in place column core.
With the increase of the diameter of cast-in-place column core, the bearing capacity and the deformation capacity of the specimen decrease.
Finally, based on the experimental research and theoretical analysis, a calculation model for predicting the maximum bearing capacity was proposed, and the results obtained from the formulas were in good agreement with the test results.
Research Data
sj-xlsx-1-ase-10.1177_13694332211011551 – Supplemental material for Experimental study on seismic performance of partially precaststeel fiber high-strength concrete columns with high-strength steel bars
Supplemental material, sj-xlsx-1-ase-10.1177_13694332211011551 for Experimental study on seismic performance of partially precaststeel fiber high-strength concrete columns with high-strength steel bars by Jianwei Zhang, Deli Zhang, Xiangyu Li and Zhaoxv Shen in Advances in Structural Engineering
Footnotes
Acknowledgements
The authors are grateful for the funding provided by the National Natural Science Foundation of China (Grant No. 51678009) and the Key Laboratory of Urban Security and Disaster Engineering MOE, Beijing University of Technology.
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 (Grant No. 51678009).
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
