Effect of recycled aggregate concrete on the seismic behavior of DfD beam-column joints under cyclic loading

Abstract

The problem of environmental pollution and waste concrete generation will be alleviated if concrete structures can be designed with both recycling and reuse strategies. In this regard, this study employed recycled coarse aggregate as the structural material in a novel beam-column joint with Design for Deconstruction (DfD) connections. Two series cyclic loading tests including pre-loading and re-loading on this frame joint was undertaken and the effect of recycled aggregate concrete (RAC) on structural behaviors was carefully evaluated. It is demonstrated that the DfD joint with RAC had favorable integrity and the influence of RAC was acceptable for practical application. Furthermore, the shear force model of the joint core area for RAC beam-column joint with DfD connections was put forward in this study, which showed good agreements with experimental results. Life cycle assessment (LCA) indicated that the carbon emission of the beam-column joint with DfD connections made of RAC increased at construction process but would reduce the total carbon emission from the perspective of lifetime.

Keywords

beam-column joints design for deconstruction (DfD)life cycle assessment (LCA)recycled aggregate concrete (RAC)shear force model

Introduction

In recent decades, large amount of concrete waste is generated with the rapid development of the building industry and natural disasters (Brown et al., 2011; Cochran et al., 2007; Ding and Xiao, 2014; UN, 2018; Xiao et al., 2012). The most usual treatment of concrete waste is landfill in rural areas, which undoubtedly has obvious adverse environmental effects, such as occupying massive land, contaminating surrounding soil environment, wasting natural resources, etc. (Tränkler et al., 1996; Yuan and Shen, 2011). On the other hand, the shortage of natural resources has become more and more prominent. It was recently reported by Nature Journal that 32 billion to 50 billion tons of sand is annually consumed worldwide, which far exceeds the natural renewal rate of river sand (Bendixen et al., 2019). Therefore, using waste concrete in a more sustainable way and reducing the consumption of natural resources have been paid much more attention worldwide (Henry and Kato, 2014; Ortiz et al., 2009; Radonjanin et al., 2013).

Currently, the 3R concept (Reduce, Reuse, and Recycle) is considered to be an important method to solve these problems (Peng et al., 1997). As for Recycle, this concept is fully embodied in those applications adopting recycled aggregate concrete (RAC). In the past few decades, sufficient experimental studies have been carried out for RAC in terms of material properties (Sagoe-Crentsil et al., 2001; Tabsh and Abdelfatah, 2009; Xiao et al., 2005), component properties (Choi and Yun, 2012; Etxeberria et al., 2007; González-Fonteboa and Martínez-Abella, 2007; Seara-Paz et al., 2018; Wardeh and Ghorbel, 2019; Xiao et al., 2014a: 692–712; Yan et al., 2018), structural properties (Liu et al., 2020; Xiao et al., 2014b: 1381–1405) and the bonding properties with reinforcement (Prince and Singh, 2013; Xiao and Falkner, 2007). Some studies also analyzed the effect of RAC on the frame beam-column joint, which is essential to the safety and integrity of recycled concrete frame structure (Corinaldesi et al., 2011; Hu and Kundu, 2019; Letelier Gonzalez and Moriconi, 2014; Xue et al., 2018). The results of previous studies proved that RAC is completely feasible as a kind of structural material through reasonable design. Many countries have already issued the specifications for the practical application of RAC (Dosho, 2007; Poon and Chan, 2007; Vandecasteele and van der Sloot, 2011).

In addition to recycled concrete, the Design for Deconstruction (DfD), a new structural design principle also meets the requirements of the 3R concept, more specifically, the requirement of Reuse. The DfD principle was firstly proposed in 2000s by CIB Task Group 39 (Kibert and Chini, 2000). The terms “Design for Deconstruction,”“Replaceable structure,” and “Demountable structure” were somewhat similar. However, “Design for Deconstruction” suggests that the components of DfD structures should have the capability of being disassembled, reinstalled and reused after their first service life. While “Replaceable structure” is a type of earthquake resilient structure developed to restore the structure immediately after strong earthquakes by replacing some components which could dissipate seismic energy (Chen et al., 2017). Moreover, the “Demountable structure” is referred to the possibility of disassembly and reinstalling of structure components, which focused more on the design of demountable connections (Uy et al., 2017).

DfD structures have already been concerned for a long time. There have been many reports on the basic principles and the design methods of DfD structures (Addis and Schouten, 2004; Akinade et al., 2017; Chini, 2001; Tingley and Davison, 2011), especially on those environmental and economic benefits. Tingley and Davison (2012) proposed one life cycle assessment (LCA) study to evaluate the environmental benefits of DfD structure. Akbarnezhad et al. (2014) developed an optimized method to evaluate the environmental and economic benefits of DfD structures using building information modeling. Eckelman et al. (2018) compared different design schemes of DfD floor system through LCA.

Obviously, reusing waste concrete components is a more sustainable method to eliminate the adverse environmental effects (Akinade et al., 2017). However, due to the requirement of structure integrity, abundant cast-in-situ concrete is always needed in DfD concrete connections, which brings difficulties to the subsequent deconstruction process and makes it necessary to conduct researches on design methods and structural performances (Ding et al., 2018). With the aim to explore the feasibility of DfD concrete structure, two series experimental tests on DfD concrete beams before deconstruction and after reconstruction were carried out by Ong et al. (2013). The revised steel plate-bolt connection, as a kind of dry connections, was used in the specimens. Recently, a novel DfD concrete connection method suitable for beam-column joints was developed by the authors (Ding et al., 2020; Xiao et al., 2017). Favorable mechanical resistance and ductility performance under static and cyclic loading of the proposed beam-column joint with ductile DfD connection was observed during the test. However, in these studies, the feasibility of RAC on DfD components and the environmental benefit were not fully analyzed and reported. Moreover, the quantitative evaluation, especially for the joint shear force in the beam-column joint, was not considered.

From the above literature reviews, it can be found that there have been sufficient reports on the environmental and economic benefits of DfD structures. The structural behaviors of DfD concrete structures have also been paid more and more attention. However, analysis on the application of RAC in DfD structures, especially in those beam-column concrete joints, has not been found yet. It seems that a higher sustainability, better environmental effects will be achieved if concrete structures can be designed with both recycling and reuse strategies. As a result, the effect of RAC on the beam-column concrete joints with DfD connections under cyclic loading were evaluated in this study. The feasibility of RAC adopted in DfD connections was explored. The shear force in the joint core area was tested and compared with current standard, a horizontal shear force model was proposed. Also, LCA based on carbon emission of the proposed RAC beam-column joint with DfD connection was conducted to confirm the improved structural sustainability with both recycling and reuse strategies.

Experimental program

Design and details of specimens

In order to assess the effect of RAC on the mechanical behavior of beam-column joint with DfD connections, a cyclic loading test on full-scale concrete beam-column joints in the plane frame was carried out. A schematic view of the joint specimen was shown in Figure 1.

Figure 1.

Dimension, reinforcement and DfD connection details of the joint specimens (Unit: mm) (Ding et al., 2020).

For the DfD connecting method in this test, the mortise-tenon was adopted. Prefabricated concrete columns were cast continuously in the elevation with short protruding beams. The overhanging I-section steel of the reusable beams was inserted into the reserved groove of the prefabricated concrete columns. The horizontal reinforcements of the beam were connected by the welding method to enhance continuity of specimens. Due to the reduction of the cross-sectional area of beam section near the connection part, the I-section and T-section steels were used to provide the adequate shear bearing capacity for beams. Finally, a small quantity of cast-in-situ concrete is poured at the connection part. For the demounting process, a small jackhammer was used to remove the concrete surrounding the welding steel bars, and then the steel bar was cut off to separate the specimens. Welding method and cast-in-situ concrete were also used when reinstalling the beam-column connection.

Specimens NJ-M and RJ-M were monolithic joints made of natural aggregate concrete (NAC) and RAC, respectively. Specimens NJ-DfD and RJ-DfD were joints with DfD connections made of NAC and RAC, respectively. In order to obtain more realistic mechanical properties and avoid the result distortion caused by size effect, full-scale specimens were prepared in this test. The length of the beam and the column was both 3200 mm. The section sizes of the beam and the column were 200 mm × 400 mm and 350 mm × 350 mm, respectively.

The design of joint specimens meets the requirements of “strong column and weak beam, strong shear and weak bending, strong joint and weak component” in GB50011-2010. Figure 1 also showed the dimension, reinforcement and connection details of joint specimens.

Materials

RAC with 100% coarse aggregate replacement and NAC were used in this test. The designed strength grade of concrete is C30. The mix design was based on a similar compressive strength of RAC and NAC, at 28 days under the standard curing condition.

Ordinary Portland cement, natural coarse aggregate (NCA), recycled coarse aggregate (RCA) with size ranged from 5 to 20 mm, river sand, tap water and polycarboxylate superplasticizer were used for the concrete mix. The basic properties of NCA and RCA were presented as follow: the apparent density was 2660 kg/m³ and 2530 kg/m³, the clay content was 0.8% and 1.1%, the water absorption was 1.0% and 3.5%, the crushing value index was 5.1% and 11%, respectively. The particle distribution of coarse recycled aggregates and natural aggregate measured through sieve analysis, which was shown in Figure 2.

Figure 2.

Particle-size distribution of coarse recycled aggregate.

With the aim of obtaining similar compressive strength and working performance, the mix proportion of NAC and RAC were adjusted by changing water-cement ratio, water consumption and sand rate. The mix proportions, presented as the ratio of the mass of cement, water, coarse aggregate, fine aggregate and water-reducing admixture, of NAC and RAC were 1:0.45:3.04:1.92:0.5% and 1:0.42:2.69:1.94:0.8%. Some additional water was added in RAC according to the water absorption property of RCA. The net water-cement ratio of NAC and RAC was 0.45 and 0.42, respectively.

Since the mechanical performances of DfD specimens were tested after reinstallation, $150 mm \times 150 mm \times 150 mm$ cube blocks were reserved during casting to obtain the concrete compressive strength at different ages. The cube blocks were tested on the test day. The test of monolithic specimens was at the ages of 60 days, while the test of DfD specimens was 2 months later than monolithic specimens. The compressive strength of concrete was 36.19 MPa, 39.74 MPa, 46.16 MPa, and 51.53 MPa for specimens NJ-M, RJ-M, NJ-DfD and RJ-DfD, respectively. The strength of concrete, both RAC and NAC, increased with the growth of age, and the strength of RAC is about 10% higher than that of NAC. This deviation was not only due to the unavoidable inaccuracy of concrete mix design but also the larger increase of compressive strength of RAC at the longer age (Kou et al., 2011). The elastic modulus of concrete specimens NJ-M, RJ-M, NJ-DfD, RJ-DfD was 32.71 GPa, 36.08 GPa, 31.82 GPa, 33.65 GPa, respectively.

The steel bars of HRB400 and HPB300 were used as longitudinal bars and stirrups in the test, respectively. The longitudinal bars at the beam section were welded by single side lap arc welding using the E50 electrode, and the weld length was 100 mm. Tensile yield strength tests were carried out on steel bars and the longitudinal welded bars. The results presented that the yield strength of steel bar was about 416 MPa. The elastic modulus of steel bar was about 190 GPa. The yield strength of welded bars was similar to that of the longitudinal bars used in the beam section, indicating that the welding connection could obtain a good force transfer performance in this test.

Test procedure

A customized test set-up was used in this test (Ding et al., 2020). Beams and columns of specimens were pinned at their ends. Each specimen was subjected to a constant axial load 405 kN on the top of the column by a hydraulic jack. Due to the limitation of test setup, the available maximum axial load of 405 kN was applied during the test, in order to maximize the actual axial load ratio. In fact, the axial load ratios of specimens were 0.12, 0.11, 0.09, and 0.08 of NJ-M, RJ-M, NJ-DfD, RJ-DfD, respectively. The axial load ratios were relatively close, which can achieve the test purpose. Horizontal cyclic loading at the top of column by an actuator, simulated the earthquake loading. The selected lateral load history consisted of load-control initially and displacement-control after beam bars yielding. The horizontal load was applied cyclically in a quasi-static way at the top of column. The beam longitudinal bars yielded at near 40 kN for four specimens, so the load 40 kN was chosen to be the end of load-control stage. At the load-control stage, each specimen was loaded by increased load of 10 kN and one cycle was applied for each step until the load reached 40 kN. The specimen displacement at 40 kN was recorded as Δ₀. At the displacement-control stage, each specimen was loaded by increased displacement of 0.5 Δ₀ and three cycles were applied for each step until the bearing capacity dropped to 85% of the peak bearing capacity, which was regarded as specimen damaged.

Two monolithic specimens NJ-M and RJ-M were directly tested up to failure, while two DfD specimens NJ-DfD and RJ-DfD were installed, pre-loaded, demounted and reinstalled before tested up to failure. The pre-load history was displacement-control by 8 mm increased for each step until the displacement reached 64 mm. The displacement of 64 mm represented the interstory drift ratio of 2%, which was regarded as the limit state of structures after an earthquake (GB50011-2010).

A thin layer of lime paste was applied to the concrete surface of each specimen, and a $50 mm \times 50 mm$ square grid was drawn to record the generation and propagation of cracks during the test. Linear Variable Displacement Transducers (LVDTs) were located to measure the displacement at column ends and the joint core area. Strain gauges were located in longitudinal bars and stirrups in the joint core area to analyze the strain distribution and deformation of specimens. Figure 3 showed the test set-up, strain gauges and LVDTs location of test specimens, where N₀ is the constant vertical load applied on the top of the column and P is the horizontal cyclic loading.

Figure 3.

Test set-up, strain gauges and LVDT locations of specimens (Ding et al., 2020).

Test results and analysis

Crack propagation and failure pattern

Specimen NJ-M and specimen RJ-M

At the load-control stage, when the load reached 20 kN, micro-cracks appeared at the beam end near the beam-column interface. After that, cracks gradually developed from top to bottom of the beam section. At the displacement-control stage, when the displacement reached 40 mm, inclined micro-cracks appeared in the joint core area. Afterwards, when the displacement reached 100 mm, oblique cracks completely passed throughout the beam section and the joint core area, no new cracks were then generated. At this stage, uniform vertical bending cracks were observed in the beam of specimen NJ-M while uneven oblique cracks were distributed in the beam of specimen RJ-M. Finally, plastic hinges appeared near the beam-column interface, a large amount of concrete peeled off at this area. The crack propagation processes of specimen NJ-M and specimen RJ-M were similar, however, the number of cracks in the joint core area of RAC specimen RJ-M was significantly larger than that of NAC specimen NJ-M.

Specimen NJ-DfD and specimen RJ-DfD

At the load-control stage, when the load reached 20 kN, micro-cracks continued to develop along the cracks generated in the pre-load stage of the DfD specimens. The development of cracks was impeded by section steels, showing a trend of inclined development. At the displacement-control stage, when the displacement reached 40 mm, inclined micro-cracks appeared in the joint core area and horizontal micro-cracks appeared at the end of column. When the displacement reached 100 mm, the oblique cracks in the joint core area stopped developing, while the oblique cracks completely passed throughout the beam and the old-to-new concrete interface. At this stage, cracks in the concrete beam section were mainly distributed along edges of section steels and old-to new concrete interface. Finally, the plastic hinges also appeared near the beam-column interface with quantities of concrete peeled off. In general, the concrete crushing at the plastic hinges of RAC specimen RJ-DfD was significantly more severe than that of NAC specimen NJ-DfD.

The failure pattern of all four specimens was plastic hinge failure near the beam-column interface. Obvious inclined crack development and shear damage in the core area was also observed, which were shown in Figures 4 and 5. It can be found that the concrete crushing and peeling in the joint core area and the plastic hinge of RAC specimens was more severe than that of NAC specimens in both monolithic and DfD specimens.

Figure 4.

Failure pattern near the beam-column interface.

Figure 5.

Crack patterns in the joint core area of specimens: (a) NJ-M, (b) RJ-M, (c) NJ-DfD, and (d) RJ-DfD.

Section steels were arranged in the beam section of DfD specimens to strengthen the shear capacities. Therefore, the internal force and the crack distribution of the beam section in DfD specimens were significantly different from those of monolithic specimens. However, it can also be observed that DfD has less impact on the crack propagation at the joint core area, mainly because the DfD connection part was far away from this area.

Hysteretic behavior

Hysteretic curves of test specimens subjected to cyclic loading were depicted in Figure 6. The overall characteristics of four hysteretic curves were basically similar, indicating that the application of RAC and demountable method has limited influence on the energy dissipation capacity of beam-column joint specimens. At the initial loading stage, a basically linear relationship between load and displacement was observed and the stiffness degradation of specimens was not pronounced. DfD specimens NJ-DfD and RJ-DfD with initial damage exhibited similar stiffness when compared to monolithic specimens. With the load increased, the area enclosed by hysteretic curves gradually increased, indicating a favorable energy dissipation. After specimens entered the nonlinear stage, the hysteretic loop presented a typical inverse Z-shape and showed somewhat pinch-effect, which was consistent with the obvious inclined crack development and shear damage in the core area of specimens.

Figure 6.

Hysteretic curves of specimens: (a) NJ-M, (b) RJ-M, (c) NJ-DfD, and (d) RJ-DfD.

Envelope curves

The envelope curves were obtained through connecting the margin points of the hysteretic curves. The characteristic points of envelope curves were calculated according to the energy equivalence method, the yield point of the joint could be defined by nominal yield moment (Yao and Chen, 2001), while the ultimate point was corresponding to the point when the lateral load dropped to 85% of the peak bearing capacity, which were listed in Table 1.

Table 1.

Characteristic points of envelope curves.

Specimens	Loading direction	$P_{y} (kN)$	$Δ_{y} (mm)$	$P_{max} (kN)$	$Δ_{max} (mm)$	$P_{u} (kN)$	$Δ_{u} (mm)$	$\frac{Δ_{u}}{Δ_{y}}$
NJ-M	+	52.72	44.8	64.5	62.4	54.83	191.04	4.26
NJ-M	–	62.42	39.68	75.13	95.36	63.83	152.64	3.85
RJ-M	+	56.84	49.6	70.75	100.8	60.14	191.04	3.85
RJ-M	–	60.74	48.96	69.26	82.56	58.87	161.92	3.31
NJ-DfD	+	67.24	56.96	71.64	79.68	60.89	145.92	2.56
NJ-DfD	–	68.54	43.84	78.98	97.92	66.78	192.96	4.4
RJ-DfD	+	65.57	54.4	75.32	98.24	64.02	154.56	2.84
RJ-DfD	–	71.57	49.28	80	99.52	68	146.56	2.97

Due to different ages of monolithic specimens NJ-M, RJ-M and DfD specimens NJ-DfD, RJ-DfD, the concrete strength was slightly different when tested. Therefore, in order to avoid the influence of different concrete strength and highlight the effects of different concrete materials and design methods, the normalized $λ$ -interstory drift ratio envelope curves were depicted. Normalized $λ$ was defined as follow:

λ = V_{j, h} / V_{n}

(1)

where $V_{j, h}$ is the experimental horizontal shear force in the joint core area calculated by shear force at the column end, which was later described in detail in equation (11). $V_{n} = 0.85 \times 0.083 γ \sqrt{{f'}_{c}} b_{j} h_{c}$ is the recommended horizontal shear force in the joint core area by ACI-ASCE Committee 352 (2002). $γ$ is the coefficient of shear strength and equals 15 in this study according to ACI-ASCE Committee 352, ${f_{c}}^{'}$ is the concrete compressive strength, $b_{j}$ is the effective width of joint and $h_{c}$ is the height of column section.

The normalized $λ$ -interstory drift ratio envelope curves for four specimens are depicted in Figure 7. Comparing NAC and RAC specimens, the envelope curves of RAC specimens were basically in coincidence with those of NAC specimens during the initial stage of loading. This is because the concrete of specimen was mainly subjected to compressive stress during the initial stage of loading, so the application of RAC had little impact on the overall performance. However, after reaching the peak load, the bearing capacity of RAC specimens was overall lower than that of NAC specimens. The bearing capacity of RAC and NAC specimens was similar in the positive direction, while the bearing capacity of RAC specimens was significantly lower than that of NAC specimens after reaching the peak load in the negative direction. The difference of the performance between the positive and negative direction was due to the initial directivity of the loading system. Comparing monolithic and DfD specimens, the envelope curves of monolithic specimens were basically in coincidence with that of DfD specimens at the beginning, indicating that the damage in pre-load process had little influence on the bearing capacity after reinstallation. However, after reaching the peak load, the descending branch of envelope curve of DfD specimens was steeper than that of monolithic specimens, which may be related to severer damage and lower deformation ability of DfD specimens.

Figure 7.

Normalized $λ$ -displacement envelope curves of specimens.

Strain analysis

Strain gauges were arranged to investigate the strain of steel bars in this test. The strain of stirrups at the joint core area were shown in Figure 8. Some strain gauges failed due to the serious damage at the joint core area during the test. The stirrup at the joint core area of specimen RJ-DfD had yielded when displacement reached 60 mm, while the stirrup at the joint core area of specimen NJ-DfD was still in the elastic stage during the test. The strain of stirrups at the joint core area of specimen RJ-DfD was obviously larger than that of specimen NJ-DfD. This is because the concrete damage at the joint core area of specimen RJ-DfD was more serious than that of specimen NJ-DfD, leading to a relatively larger shear stress of stirrups at the joint core area. Therefore, more attention should be paid to the stirrup layout and the shear capacity in the design of DfD concrete joint made of RAC.

Figure 8.

Strain behavior of stirrups in the joint core area.

The strain distribution of beam longitudinal bars was also monitored, shown in Figure 9. In DfD specimens, the distance of 1.0 h, 1.5 h from the beam-column interface, where h is the height of the beam, was the position near cast-in-situ concrete in overhanging short beam and in demountable beam, respectively. A sudden change of strain was not observed during the whole test, indicating that the strain continuity of DfD specimens was both good in NAC and RAC specimens. The yield strain of steel bars was 2257 με, and only the longitudinal bars at the beam-column interface were yielded, which was consistent with the plastic hinges observed during the test.

Figure 9.

Strain behavior of beam bottom longitudinal bars: (a) NAC specimens and (b) RAC specimens.

The strain of steel bars at 0 h of DfD specimens developed more slowly than that of monolithic specimens, while the strains of steel bars at 1.0 h, 1.5 h of DfD specimens were slightly larger than that of monolithic specimens. Moreover, the magnitude of the steel strain 1.0 h and 1.5 h showed an opposite trend between monolithic specimens and DfD specimens.

Overall, the strain of steel bars near the beam-column interface decreased while the strain of steel bars away from the beam-column interface increased in DfD specimens, compared to monolithic specimens. This may because of stress redistribution and outward move of plastic hinges due to the existence of section steels and cast-in situ concrete in DfD specimens. However, this phenomenon of RAC specimens was not so obvious since serious damage occurred at the joint core area of RAC specimens.

Moreover, the strain of the reinforcement of column in the core area of four specimens was observed. The test results of the four specimens were basically the same, which showed that the reinforcement of column was in the elastic stage in the test, due to the rational design of “Strong column and weak beam.”

Stiffness degradation

Secant stiffness under the same deformation was used to quantitatively describe the stiffness of specimens, which is the ratio of the sum of the absolute value of positive and negative peak load to the sum of the absolute value of the corresponding displacement in each cycle.

K_{i} = \frac{| + P_{i} | + | - P_{i} |}{| + u_{i} | + | - u_{i} |}

(2)

Where, $K_{i}$ is the stiffness in the i_th loading (or displacement) cycle, $\pm P_{i}$ , $\pm u_{i}$ are positive (negative) peak load and corresponding displacement in the i_th loading (or displacement) cycle.

All four specimens showed obvious stiffness degradation, which was depicted in Figure 10. The stiffness degradation of each specimen was fast in the initial stage of loading due to the cracking and damage propagation. However, at the late stage of loading, cracking propagation almost stopped, and the stiffness degradation gradually slowed down and finally became stable. In general, the initial stiffness of NAC specimens was larger than that of RAC specimens and the stiffness degradation was also more obvious at the beginning. However, their stiffness at the late stage of loading tended to be consistent. The stiffness degradation of DfD specimens was slightly different from that of monolithic specimens. It is shown that the inflection point between the descent part and the smoothing part appeared earlier, and the stiffness of DfD specimens was somewhat higher than that of monolithic specimens at the late stage of loading.

Figure 10.

Stiffness degradation.

Energy dissipation

The energy dissipation capacity is usually calculated by the area of hysteretic loops. The different concrete strength should be considered reasonably in the analysis of energy dissipation because it has a great influence.

The energy dissipation of the first cycle of each loading stage was shown in Figure 11. As can be seen from the figure, at the initial loading stage, four curves were well consistent with each other, indicating that the energy dissipation capacity of four specimens was similar at the beginning. After specimens entered the elastic-plastic stage, the energy dissipation of DfD specimens were higher than that of monolithic specimens. Due to the lower tensile strength of RAC, the crack propagation was more serious and the energy dissipation was reduced at the late loading stage. As a consequence, although the concrete compressive strength of specimen RJ-M was higher than that of specimen NJ-M, the energy dissipation capacity of specimen RJ-M was lower than that of specimen NJ-M, which agreed well with the more significant pinch-effect of hysteretic curves for RAC specimens. However, the effect of RAC on joint specimens with DfD connection is not obvious.

Figure 11.

Scale energy-drift ratio relation.

Damage index

The Park and Ang (1985) damage model was applied to analyze the seismic structural damage, which was expressed as a linear combination of the damage caused by excessive deformation and repeated cyclic loading effect. Damage index $D$ was defined as follow:

D = \frac{δ_{M}}{δ_{u}} + \frac{β}{Q_{y} δ_{u}} \int dE

(3)

where, $δ_{M}$ is the maximum deformation within the considered cycle, $δ_{u}$ is the ultimate deformation under monotonic loading, $Q_{y}$ is the calculated yield strength, $β$ is a non-negative strength degradation parameter, $\int dE$ means the cumulative energy dissipation of considered cycle. Due to the insufficient data, the $Q_{y}$ and $δ_{u}$ were estimated by $P_{y}$ and $Δ_{u}$ from the envelope curves of specimens (shown in Table 1) (Sharma and Bansal, 2019), the $β$ was taken as 0.1 in this study (Altoontash, 2004). The calculated results were depicted in Figure 12. The damage indexes of four specimens were similar at the beginning. However, after the displacement of 50 mm, the damage indexes of DfD specimens were obviously larger than that of monolithic specimens. Probably due to the poorer ductility and higher energy consumption in elastic-plastic loading stage of joint specimens with DfD connection. Moreover, the significant discrepancy was not observed between damage indexes of RAC and NAC joint with monolithic and DfD connection. It can be observed that the damage indexes of RAC specimens were a little lower than those of NAC specimens, which showed that the application of RAC did not obviously accelerate the joint damage.

Figure 12.

Damage index.

Horizontal design shear force model

The brittle shear failure in the core area of joint is unexpected and should be avoided. Therefore, it is important to ensure the adequate shear bearing capacity of joints for structural design. As mentioned above, although the failure mode of specimens was not dominated by shear failure, obvious inclined cracks and shear damage in the core area could be observed during the tests. The peak shear force model could provide the reference of shear requirement of the DfD joint, as a lower limitation for the design of shear resistance.

The force diagram of the beam-column joint was shown in Figure 13. The equilibrium of the joint requires the following equation:

V_{j, h} = T_{b 1} + C_{b 2} - V_{col}

(4)

\sum M_{c} = \sum M_{b}

(5)

\sum M_{b} = (T_{b 1} + C_{b 2}) \cdot h_{b 0}

(6)

\sum M_{c} = V_{col} \cdot (H - h_{b})

(7)

where $V_{j, h}$ was horizontal shear force of the joint core area, $T_{b 1}$ and $C_{b 2}$ were the tension force and compression force of the reinforcement, respectively; $V_{col}$ is the horizontal column force; $\sum M_{c}$ and $\sum M_{b}$ was the sum of the moments at the column end and the beam end, respectively; $h_{b}$ and $h_{b 0}$ were the height and the effective height of the section of beam; H was the height of the column.

Figure 13.

Force diagram of the beam-column joint.

According to ACI-ASCE 352R, the design shear force $V_{u}$ can be calculated as the follow equation:

V_{u} = T_{b 1} + C_{b 2} - V_{col}

(8)

T_{b 1} = α f_{y} A_{s 1}

(9)

C_{b 2} = α f_{y} A_{s 2}

(10)

where the $α$ was the stress multiplier for longitudinal reinforcement, which is recommended as 1.25 in this case; $f_{y}$ was the yield strength of reinforcement; $A_{s 1}$ and $A_{s 2}$ were the area of top reinforcement and bottom reinforcement in the beam, respectively.

Combining equations (4)–(10), the following equation can be yielded:

V_{j, h} = (\frac{H - h_{b}}{h_{b 0}} - 1) \cdot V_{col}

(11)

V_{u} = α (1 - \frac{h_{b 0}}{H - h_{b}}) \cdot f_{y} (A_{s 1} + A_{s 2})

(12)

The experimental horizontal shear force and the design shear force recommended by ACI-ASCE 352R of the joint core area can be calculated by equations (11) and (12), respectively. It is obvious that the design shear force $V_{u}$ calculated from equation (12) cannot reflect the influence of RAC and DfD, which could be revised to predict the shear force of RAC beam-column joint with DfD connections.

Therefore, considering the influence of the application of RAC and DfD, based on the test results and referring to the equations of shear bearing capacity and design shear force in ACI-ASCE 352R, the predicted shear force at the core area of RAC beam-column joint with DfD connections was put forward as follow:

\begin{matrix} V_{j, c} = 0.6 η_{c} \sqrt{{f'}_{c}} b_{j} h_{c} + 0.4 α η_{d} (1 - \frac{h_{b 0}}{H - h_{b}}) \\ \cdot f_{y} (A_{s 1} + A_{s 2}) \end{matrix}

(13)

where $V_{j, c}$ is the calculated modified shear force, the first part of the equation can be considered as the contribution of concrete strength and joint dimension, $η_{c}$ is the concrete material coefficient. For NAC specimens, $η_{c} = 1.0$ ; for RAC monolithic specimens, $η_{c} = 0.8$ ; for RAC beam-column joint with DfD connections, $η_{c} = 0.95$ . ${f_{c}}^{'}$ is the designed compressive strength of concrete, $b_{j}$ is the effective joint width and $h_{c}$ is the depth of column in the direction of joint shear according to ACI-ASCE 352R. The second part of the equation can be considered as the contribution of bending moment of the beam, the section steels in DfD joint should be neglected in the calculation of it. $η_{d}$ is the demountable coefficient, for monolithic specimens, $η_{d} = 1.0$ ; for DfD specimens, $η_{d} = 0.95$ .

The results of experimental shear force $V_{j, h}$ , the calculated shear force $V_{u}$ of equation (12) and $V_{j, c}$ of equation (13) were shown in Figure 14. It can be seen that the design shear force recommended by ACI-ASCE 352R was conservative while the proposed shear force model was in good agreement with experimental results within 10% discrepancy. However, due to the limited experimental data, there would be some experimental error occurred which make the value of η_c and η_d not fully reliable. The data would be further verified and revised not only by more experimental tests, but also by the finite element modeling with variable parameters in the future.

Figure 14.

Comparison of test and calculated results.

Discussion

Feasibility of RAC application

It has been stated that the feasibility of applying RAC in DfD structures, especially in those beam-column concrete joints, have not been found. This issue has been fully emphasized in the present study based on experimental analysis. According to the results and analysis, it is found that the applications of RAC and DfD both have influences on mechanical behavior of joint specimens, but the influence mechanism was quite different.

The influence of RAC on the mechanical performances mainly occurred at the late loading stage. At the initial stage, concrete was mainly subjected to compressive load, so there was no obvious difference between the specimens made of NAC and RAC. Therefore, the hysteretic curve and the envelope curves at the initial loading stage were also similar. However, at the late loading stage, especially after reaching peak load, the effects of the low tensile strength of RAC and the lower bond property between RAC and steel bars became more obvious due to the complex stress state of concrete. But only little impact on the overall structure performance have been found because the reinforcement bars resisted part of the external load. More obvious pinch-effect of hysteretic curve, slightly lower loading capacity and energy dissipation capacity of specimen RJ-DfD than specimen NJ-DfD were observed. Crack propagation and concrete crushing in the joint core area and the plastic hinge area of specimen RJ-DfD was more serious when specimen damaged. Stirrups at the joint core area of specimen RJ-DfD were yielded while stirrups at the joint core area of specimen NJ-DfD were still in elastic stage during the test.

While for the DfD specimen adopting material of RAC, the applications of both RAC and DfD didn’t cause a significant deterioration of joint specimens. On the contrary, compared to monolithic specimens, the effect of RAC on DfD specimens were even less obvious. For example, the maximum bearing force ( $λ$ ) of specimen RJ-M was 10.1% lower than that of specimen NJ-M, while that of specimen RJ-DfD was only 4.1% lower than that of specimen NJ-DfD. Energy dissipation capacity of RJ-DfD was similar to that of specimen NJ-DfD, while the energy dissipation capacity of specimen RJ-M was obviously lower than that of specimen NJ-M. This probably because that section steels were used in beam section of DfD specimens, which bore a part of stress and diminished the influence of concrete properties on the joint specimens and also indicates that DfD concrete joint made of RAC is applicable.

Environmental assessment

In order to assess how the structural sustainability will be improved by designing with both recycling and reuse strategies, the environmental benefit of the proposed RAC beam-column joint with DfD connection was evaluated by comparable LCA based on carbon emission.

The system boundary was cradle-to-gate. Therefore, the production and transportation of raw construction materials were fully considered. The data of transport distance and carbon emission were referred to local conditions. The detailed analysis of these data can be seen in the literature (Xiao et al., 2018), which will not be explained here. It is assumed that the designed service life of demountable beam was twice as that of precast column, which means in the second service life of DfD specimens, only precast column should be re-made. The joint specimens were assumed not to be used in aggressive environmental conditions, therefore, the durability problem of RAC would not be considered in this study.

The LCA results were shown in Figure 15. In the first service life, the application of DfD increased nearly 7% of the carbon emission, due to the use of section steels in DfD connections. However, about 20% reduction was observed of DfD specimens in the second service life and the total carbon emission was also about 7% lower than that of monolithic specimens. Indicating that reasonable DfD will increase the carbon emission at the first construction process, but it is obvious that the environmental benefit will be gained in the whole process when considering the reuse of demountable opponents, which is consistent with the researches of Akbarnezhad et al. (2014) and Eckelman et al. (2018). Compared with NAC joints, the application of RAC obtained about 7% reduction of carbon emission on both monolithic and DfD joints. Moreover, the RAC beam-column joint with DfD connection reduced 13% carbon emission compared with monolithic NAC joint, which proved that the combination of recycling and reuse strategies could achieve better environmental benefits and improve the sustainability of concrete structure.

Figure 15.

Carbon emission of specimens: CO₂-eq (kg).

Although in this novel kind of beam-column concrete joints, only demountable beam can be reused while the precast beam should be demolished andre-made in every service life, obvious improvement of environmental benefits was observed, especially when combining the applications of DfD and RAC. Therefore, better environmental effects will be gained if the proportion of reusable components is increased in the design of DfD structure in further studies.

Conclusion

In order to improve the sustainability of concrete structures and explore the feasibility of using RAC in DfD concrete structures, a series of cyclic loading tests on novel DfD beam-column concrete joints made of RAC was undertaken in this study. The structural behaviors were analyzed and a horizontal design shear force model was put forward. Moreover, the enviro nmental assessment was conducted to evaluate the environment effect. Based on the test results and analysis, the following conclusions can be drawn:

The failure pattern of all specimens was plastic hinge failure at the beam end. The applications of recycling strategy RAC and reuse strategy DfD both affected structural behaviors of joint specimens, but the influence mechanism was quite different. The application of RAC has slightly adverse effect on DfD joint specimens, especially in the late loading stage. However, it is found that compared to monolithic specimens, the effect of RAC on DfD specimens were even less obvious. Generally, DfD concrete joint made of RAC is applicable.

A horizontal design shear force model of the joint core area was put forward. The effect of the application of RAC and DfD were considered carefully in the proposed shear force model, which showed good agreement with experimental results.

The results of LCA showed that the proposed beam-column joint with DfD connections would increase the carbon emission at the first construction process but reduce total carbon emission, considering the reuse of demountable components. Moreover, the RAC beam-column joint with DfD connections had the lowest carbon emission, reduced about 13% carbon emission compared with conventional joint.

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 investigation was based in part upon work supported by the National Natural Science Foundation of China under Grant No. 51808399 and the joint research project between NSFC and PSF under Grant No. 51661145023.

ORCID iDs

Jianzhuang Xiao

Tao Ding

References

ACI-ASCE Committee 352 (2002) Recommendations for design of beam-column connections in monolithic reinforced concrete structures.

Addis Schouten (2004) Principles of design for deconstruction to facilitate reuse and recycling. CIRIA.

Akbarnezhad

Ong

KCG

Chandra

(2014) Economic and environmental assessment of deconstruction strategies using building information modeling. Automation in Construction 37: 131–144.

Akinade

Oyedele

Ajayi

, et al. (2017) Design for Deconstruction (DfD): Critical success factors for diverting end-of-life waste from landfills. Waste Management 60: 3–13.

Altoontash

(2004) Simulation and damage models for performance assessment of reinforced concrete beam-column joints. Doctoral Dissertation, Department of Civil and Environmental Engineering, Stanford University, CA.

Bendixen

Best

Hackney

, et al. (2019) Time is running out for sand. Nature 571(7763): 29–31.

Brown

Milke

Seville

(2011) Disaster waste management: A review article. Waste Management 31(6): 1085–1098.

Chen

Xiao

Chen

(2017) Study on the shear wall structure with combined form of replaceable devices. Advances in Structural Engineering 21(9): 1327–1348.

Chini

(2001) Deconstruction and materials reuse: Technology, economic, and policy. CIB Report, Task Group, 39.

10.

Choi

Yun

(2012) Compressive behavior of reinforced concrete columns with recycled aggregate under uniaxial loading. Engineering Structures 41: 285–293.

11.

Cochran

Townsend

Reinhart

, et al. (2007) Estimation of regional building-related C&D debris generation and composition: Case study for Florida, US. Waste Management 27(7): 921–931.

12.

Corinaldesi

Letelier

Moriconi

(2011) Behaviour of beam–column joints made of recycled-aggregate concrete under cyclic loading. Construction and Building Materials 25(4): 1877–1882.

13.

Ding

Xiao

(2014) Estimation of building-related construction and demolition waste in Shanghai. Waste Management 34(11): 2327–2334.

14.

Ding

Xiao

Chen

, et al. (2020) Experimental study of the seismic performance of concrete beam-column frame joints with DfD connections. Journal of Structural Engineering 146(4): 4020036.

15.

Ding

Xiao

Zhang

, et al. (2018) Experimental and numerical studies on design for deconstruction concrete connections: An overview. Advances in Structural Engineering 21(14): 2198–2214.

16.

Dosho

(2007) Development of a sustainable concrete waste recycling system - Application of recycled aggregate concrete produced by aggregate replacing method. Journal of Advanced Concrete Technology 5(1): 27–42.

17.

Eckelman

Brown

Troup

, et al. (2018) Life cycle energy and environmental benefits of novel design-for-deconstruction structural systems in steel buildings. Building and Environment 143: 421–430.

18.

Etxeberria

Marí

Vázquez

(2007) Recycled aggregate concrete as structural material. Materials and Structures 40(5): 529–541.

19.

GB50011-2010 (2010) Code for Seismic Design of Buildings. Beijing: China Architecture & Building Press.

20.

González-Fonteboa

Martínez-Abella

(2007) Shear strength of recycled concrete beams. Construction and Building Materials 21(4): 887–893.

21.

Henry

Kato

(2014) Sustainable concrete in Asia: Approaches and barriers considering regional context. Construction and Building Materials 67: 399–404.

22.

Kundu

(2019) Seismic performance of interior and exterior beam-column joints in recycled aggregate concrete frames. Journal of Structural Engineering 145(3): 4018262.

23.

Kibert

Chini

(2000) Overview of deconstruction in selected countries. CIB Report, Task Group, 39.

24.

Kou

Poon

Etxeberria

(2011) Influence of recycled aggregates on long term mechanical properties and pore size distribution of concrete. Cement and Concrete Composites 33(2): 286–291.

25.

Letelier Gonzalez

Moriconi

(2014) The influence of recycled concrete aggregates on the behavior of beam-column joints under cyclic loading. Engineering Structures 60: 148–154.

26.

Liu

Xue

, et al. (2020) Experimental study on seismic performance of steel-reinforced recycled concrete frame with infill wall. The Structural Design of Tall and Special Buildings 29: e1744.

27.

Ong

KCG

Lin

Chandra

, et al. (2013) Experimental investigation of a DfD moment-resisting beam–column connection. Engineering Structures 56: 1676–1683.

28.

Ortiz

Castells

Sonnemann

(2009) Sustainability in the construction industry: A review of recent developments based on LCA. Construction and Building Materials 23(1): 28–39.

29.

Park

Ang

AHS

(1985) Mechanistic seismic damage model for reinforced concrete. Journal of Structural Engineering 111(4): 722–739.

30.

Peng

Scorpio

Kibert

(1997) Strategies for successful construction and demolition waste recycling operations. Construction Management and Economics 15(1): 49–58.

31.

Poon

Chan

(2007) The use of recycled aggregate in concrete in Hong Kong. Resources, Conservation and Recycling 50(3): 293–305.

32.

Prince

MJR

Singh

(2013) Bond behaviour of deformed steel bars embedded in recycled aggregate concrete. Construction and Building Materials 49: 852–862.

33.

Radonjanin

Malešev

Marinković

, et al. (2013) Green recycled aggregate concrete. Construction and Building Materials 47: 1503–1511.

34.

Sagoe-Crentsil

Brown

Taylor

(2001) Performance of concrete made with commercially produced coarse recycled concrete aggregate. Cement and Concrete Research 31(5): 707–712.

35.

Seara-Paz

González-Fonteboa

Martínez-Abella

, et al. (2018) Flexural performance of reinforced concrete beams made with recycled concrete coarse aggregate. Engineering Structures 156: 32–45.

36.

Sharma

Bansal

(2019) Behavior of RC exterior beam column joint retrofitted using UHP-HFRC. Construction and Building Materials 195: 376–389.

37.

Tabsh

Abdelfatah

(2009) Influence of recycled concrete aggregates on strength properties of concrete. Construction and Building Materials 23(2): 1163–1167.

38.

Tingley

Davison

(2011) Design for deconstruction and material reuse. Proceedings of the Institution of Civil Engineers - Energy 164(4): 195–204.

39.

Tingley

Davison

(2012) Developing an LCA methodology to account for the environmental benefits of design for deconstruction. Building and Environment 57: 387–395.

40.

Tränkler

JOV

Walker

Dohmann

(1996) Environmental impact of demolition waste - An overview on 10 years of research and experience. Waste Management 16(1): 21–26.

41.

UN (2018) World Urbanization Prospects: The 2018 Revision, Custom Data Acquired Via Website. United Nations (UN), Department of Economic and Social Affairs, Population Division.

42.

Patel

, et al. (2017) Behaviour and design of connections for demountable steel and composite structures. Structures 9: 1–12.

43.

Vandecasteele

van der Sloot

(2011) Sustainable management of waste and recycled materials in construction. Waste Management 31(2): 199–200.

44.

Wardeh

Ghorbel

(2019) Shear strength of reinforced concrete beams with recycled aggregates. Advances in Structural Engineering 22(8): 1938–1951.

45.

Xiao

Ding

Zhang

(2017) Structural behavior of a new moment-resisting DfD concrete connection. Engineering Structures 132: 1–13.

46.

Xiao

Falkner

(2007) Bond behaviour between recycled aggregate concrete and steel rebars. Construction and Building Materials 21(2): 395–401.

47.

Xiao

Zhang

(2005) Mechanical properties of recycled aggregate concrete under uniaxial loading. Cement and Concrete Research 35(6): 1187–1194.

48.

Xiao

Wang

Ding

, et al. (2018) A recycled aggregate concrete high-rise building: Structural performance and embodied carbon footprint. Journal of Cleaner Production 199: 868–881.

49.

Xiao

Xie

Zhang

(2012) Investigation on building waste and reclaim in Wenchuan earthquake disaster area. Resources, Conservation and Recycling 61: 109–117.

50.

Xiao

Pham

Wang

, et al. (2014a) Behaviors of semi-precast beam made of recycled aggregate concrete. The Structural Design of Tall and Special Buildings 23(9): 692–712.

51.

Xiao

Wang

Pham

, et al. (2014b) Nonlinear analysis and test validation on seismic performance of a recycled aggregate concrete space frame. The Structural Design of Tall and Special Buildings 23(18): 1381–1405.

52.

Xue

Zhai

Bao

, et al. (2018) Cyclic loading tests and shear strength of steel reinforced recycled concrete beam-column inner joints. The Structural Design of Tall and Special Buildings 27(6): e1454.

53.

Yan

Liu

Zou

, et al. (2018) Eccentric compressive behavior of recycled aggregate concrete columns after freezing and thawing cycles. Advances in Structural Engineering 21(15): 2299–2310.

54.

Yao

Chen

(2001) Civil Engineering Structure Test. Beijing: China Architecture & Building Press. [In Chinese]

55.

Yuan

Shen

(2011) Trend of the research on construction and demolition waste management. Waste Management 31(4): 670–679.