Mechanical behaviors of GFRP tube confined recycled aggregate concrete with sea sand

Abstract

An experimental program was undertaken to study the mechanical behaviors of glass fiber-reinforced polymer (GFRP) tube confined recycled aggregate concrete with sea sand (GRACSS) under the axial compression. Two different parameters were mainly considered: recycled coarse aggregates (RCA) replacement percentage (0, 100%) and type of sand (sea sand, river sand). Typical influences of RCA and sea sand on the strength, the deformation and the load–deformation curve of GRACSS were investigated. The test results showed that the failure pattern of GRACSS was similar to that of GFRP tube confined ordinary concrete (GCOC). The strength of GRACSS decreased with an increasing RCA replacement percentage, while sea sand could reduce the negative effect of RCA. It is also found that the peak deformation of GRACSS increased with the increasing RCA replacement percentage whereas with decreasing sea sand chloride ion (Cl^–) content. The stiffness of the specimen was obviously influenced by the concrete type. Research findings indicated that the axial load-deformation curve of GRACSS can be divided into elastic-plastic and hardening stages. An analytical expression was proposed to calculate the load-deformation curve of GRACSS. Finally, the finite element method (FEM) was applied to study the effects of outer tube thickness, concrete strength, RCA replacement percentage and Cl^– content in sea sand on the mechanical behaviors (strength and deformation) of GRACSS.

Keywords

glass fiber-reinforced polymer recycled aggregate concrete recycled coarse aggregate sea sand mechanical behavior

Introduction

The rapid development of construction engineering has brought serious problems such as environmental pollution and excessive resource consumption. It was found that the annual consumption of river sand and gravel was huge and the output of construction waste (mortar, stone, bricks, etc.) increased rapidly (Huang et al., 2020; Kourmpanis et al., 2008). Thus, it is important to find alternative materials and develop eco-friendly concrete.

Recycled coarse aggregate concrete with sea sand (RACSS) is considered as an ideal kind of materials. The river sand and gravel can be partly or totally replaced by sea sand and recycled coarse aggregates (RCA), respectively, in the production of RACSS. It is an important type of eco-friendly concrete. Sea sand has many advantages such as low clay content, desired size and low cost (Limeira et al., 2011; Teng et al., 2019), and the typical properties of RCA are similar to those of natural coarse aggregates (NCA) (Li et al., 2017; Xiao et al., 2012). The application of RACSS in engineering is promising (Huang et al., 2018). However, RCA contain many initial defects and older mortar, and there are high chloride ions (Cl⁻) and shell particles contents in sea sand (Huang et al., 2018). The mechanical properties of RACSS are complicated due to the above-mentioned issues.

At present, there are few studies on the mechanical properties of RACSS. These investigations mainly analyze the coupled effects of RCA replacement percentage and sea sand Cl⁻ content on the properties (workability, strength and deformation) of RACSS (Huang et al., 2018). Xiao et al. (2018), Zhang et al. (2019) and Kumar et al. (2016) reported that the low clay content and desired size fractions of sea sand reduced the negative influence of RCA on the workability of concrete (Bravo et al., 2015; Poon et al., 2009). The workability of RACSS could meet the requirements of engineering. It was also shown that the elastic modulus of RACSS increased with the increasing sea sand content whereas with decreasing RCA replacement percentage (Huang et al., 2020). The cubic and prismatic compressive strength of RACSS was enhanced compared to recycled aggregate concrete (RAC), and RCA reduced the influence of sea sand on the concrete strength (Huang et al., 2020; Xiao et al., 2018). It was found that the compressive strength of RACSS was higher than that of ordinary concrete (concrete adopting river sand and gravel) when Cl^– content was more than a certain value (Huang et al., 2018). This can be mainly attributed to the coupled effects of the Cl^– in sea sand and the older mortar in RCA. The Cl^– slightly increased concrete density whilst reduced its porosity and absorption capacity (Etxeberria et al., 2016; Li et al., 2015; Shi et al., 2015). The more the Cl^– content in sand, the denser the micro-structure of concrete, and the higher the compressive strength of RACSS (Etxeberria et al., 2016). Therefore, sea sand could offset the negative influence of RCA on the concrete strength when sea sand Cl^– content is more than a certain value. The tensile strength of concrete with sea sand and RCA was averagely 10% lower than that of ordinary concrete and 5% higher than that of recycled concrete (Cheng et al., 2017). This is due to the coupled effects of sea sand and RCA on the micro-and macro-structures of RACSS. Based on the above research results, it can be concluded that the mechanical properties of RACSS are acceptable. However, the corrosion of reinforced concrete structures with sea sand was serious. Dias et al. (2008), Choi et al. (2014) and Dolage (2013) reported that the corrosion of steel bars in concrete with allowable sea sand Cl⁻ content was similar to that of ordinary concrete; however, concrete with high sea sand Cl⁻ content showed obvious steel corrosion. The application of RACSS is thus somewhat restricted.

GFRP tube confined concrete is considered as an effective way of improving mechanical properties and durability of RACSS. It was found that the shrinkage of core concrete was lower than that of plain concrete due to the restraint of outer tube (Geng et al., 2015). Furthermore, the strength and the deformability of specimen were significantly improved, which can be attributed to the confinement of GFRP tube (Li et al., 2015). It should also be noted that the outer GFRP tube can be used to form a closed shell, which would prevent outside air and water penetrating into the core concrete. The corrosion of steel bars in RACSS is effectively prevented. Therefore, GFRP tube confined recycled aggregate concrete with sea sand (GRACSS) is thus a good method to extend the practical application of RACSS in construction engineering. It can save the increasingly scarce terrestrial resources, and make full use of the sufficient marine materials. However, there are few studies on the mechanical behavior of GRACSS under the axial compression (Li et al., 2016a, 2016b). Further experimental investigation and simulation analyses need to be undertaken.

This study would provide deeper insight into the behavior of GRACSS. Four groups of specimen under the axial compression were performed in the test. Two different parameters were mainly considered: RCA replacement percentage (0, 100%) and type of sand (sea sand, river sand). In addition, the numerical simulations were performed by the finite element method (FEM) to study the mechanical mechanism of the GRACSS under different tube thicknesses, concrete strength, RCA replacement percentages and sea sand Cl^– contents. Finally, analytical expressions for the peak load and the axial stress-strain cure were suggested. The obtained results would provide a basis for the practical application of GRACSS.

Experimental program

Raw materials

Two different types of coarse aggregates were adopted, that is, recycled coarse aggregates (RCA) and natural coarse aggregates (gravel) (Figure 1). RCA were obtained after crushing and sieving waste concrete from a demolished pavement. The properties of coarse aggregates (Table 1) were obtained in accordance with related Chinese standards (MCPRC, 2010, 2011a). River sand and sea sand were adopted in this test (Figure 2). The sea sand was derived from Lingshan Bay, Qingdao, China. Typical properties of sea sand and river sand are listed in Table 2 (MCPRC, 2011b). It was found that the Cl^– content of sea sand were more than those of river sand. In addition, 42.5R Portland cement was used.

Figure 1.

Coarse aggregates: (a) recycled coarse aggregates and (b) natural coarse aggregates.

Table 1.

Basic properties of coarse aggregates.

	Size/mm	Bulk density/kg·m^–3	Apparent density/kg·m^–3	Water absorption/%	Crushing index/%
RCA	5.0–31.5	1416	2500	6.9	11.73
NCA	5.0–31.5	1562	2580	1.2	8.82

Figure 2.

Fine aggregates: (a) river sand and (b) sea sand.

Table 2.

The properties of fine aggregates.

Sand	Clay content/%	Apparent density/kg·m^–3	Size/mm	Shell content/%	Cl⁻ content/%
Sea sand	0.78	2570	0.15–4.75	1.13	0.1488
River sand	2.94	2612	0.15–4.75	0	0

The target concrete strength in the study was C40. The actual mix proportion was cement: sand: coarse aggregate: water = 1: 0.71: 1.67: 0.3. There were 18 prismatic (100 × 100 × 300 mm³) and 18 cubic (100×100×100 mm³) specimens prepared to obtain the mechanical properties of concrete (f_c and f_cu).

All RCASS were cured under standard condition (room temperature of 20 ± 2°C, relative humidity of 90 ± 5%). The mechanical properties of RACSS are listed in Table 3. It was found that the compressive strength and the elastic modulus (E_c) of RAC were low compared to ordinary concrete; however, sea sand could offset the negative of RCA on the properties of concrete due the high Cl^– content.

Table 3.

Summary of results.

Specimen	RCA replacement percentage	f _c/MPa	E _c/GPa	N _max/kN	ε _cc/×10^–3ε	μ	f _cc/MPa	N_p _max/kN	N _max/N_p_max
GRACSS-0-R	0	42.3	34.22	4091	30.4	2.79	130.2	4088	1.00
GRACSS-100-R	100%	40.2	30.63	3994	32.5	2.86	127.2	3877	1.03
GRACSS-0-S	0	42.8	35.28	4062	25.6	2.74	129.3	4102	0.98
GRACSS-100-S	100%	41.0	29.40	3822	28.7	2.69	121.7	3886	0.98

Glass fiber reinforced plastic (GFRP) was made of glass fiber whose substrate is unsaturated polyester resin. The height, inner diameter and thickness of GFRP tube were 600 mm, 200 mm, and 5 mm, respectively. The hoop tensile strength of tube was 600 MPa. The reasons for choosing GFRP instead of CFRP as outer tube material are its advantages, that is, low cost and acceptable durability (Kato and Yamaguchi, 1997; Li et al., 2013).

Specimen design

There were totally four groups of specimen with different types of fine aggregates (sea sand, river sand) and RCA replacement percentages (0, 100%). Each group had two specimens. Details of specimens are listed in Table 3. The specimen was denoted by “GRACSS-RCA replacement percentage-fine aggregate type.” Taking GRACSS-100-S as an example, “GRACSS” represents GFRP tube confined recycled aggregate concrete with sea sand, “100” denotes 100% RCA replacement percentage and “S” is sea sand.

Test setup, instrumentation, and loading protocol

The loading setup consisted of two parts: a 5000 kN electro hydraulic servo tester and a computer system. Details of loading setup are shown in Figure 3. There were eight and four strain gauges positioned in the middle and bottom of GRACSS specimens, respectively. The axial deformation was measured by displacement transducers (LVDTs) which fixed in the middle span of specimen. The deformation and axial load were automatically recorded by a computer system.

Figure 3.

Loading system.

Experiment was carried out in the Building Structures Laboratory of Shandong University of Science and Technology. The loading mode was in form of displacement pattern, and the actual loading rate was 1 mm/min. In addition, all GRACSS specimens were preloaded with 50 kN before the actual loading, which can ensure the normal work of devices. The typical failure pattern was also recorded at the end of test.

Experimental results

Failure process

The typical failure process of GRACSS was similar to that of GFRP tube confined ordinary concrete (GCOC). Taking GRACSS-100-S as an example, the deformation of specimen was small at the early stage of loading. The axial deformation linearly increased with the load. The relationship between axial load (N) and deformation became nonlinear when N approached 70% to 80% of the peak load (N_max). A few white patches were observed along the fiber directions on the GFRP tube. This indicated that large scale fiber damage has happened inside of GFRP tube due to the transverse deformation of core concrete was larger than that of GFRP tube. The damage (white patches) increased with an increase in load. It was found that many white patches extended to a circle around the specimen when N reached 95% to 100% of N_max. A few white patches were locally broken and N dropped suddenly after the peak point. Typical failure patterns of GRACSS were shown in Figure 4. It can be found that the influences of sea sand and RCA on the failure pattern were not obvious.

Figure 4.

Typical failure pattern of specimen: (a) GRACSS-0-R, (b) GRACSS-100-R, (c) GRACSS-0-S, and (d) GRACSS-100-S.

Axial load-deformation curve

The axial load-deformation curve is always adopted to describe the change of mechanical behaviors of specimen. Typical axial load-strain curves of GRACSS specimens under the compression are shown in Figure 5 (the strain is the ratio of axial displacement to mid-span length of specimen (L = 400 mm)). Test results indicated that the curve of specimen with sea sand and RCA was similar to that of GCOC. The curve was mainly divided into two stages: elastic-plastic stage and hardening stage. Generally, the relation between axial load and deformation can be characterized by nonlinear at the elastic-plastic stage; however, the load almost linearly increased with the increasing deformation at the hardening stage. Furthermore, there was also a transition zone between the two stages. Sea sand and RCA slightly changed the shape of axial load-deformation curve.

Figure 5.

Axial load strain of specimen.

Test results indicated that the initial stiffness of GRACSS increased with a decrease in RCA replacement percentage due to many initial defects and older mortar in RCA. However, the influence of sea sand on the initial stiffness was positive because of sea sand improved the micro-structure of concrete (Li et al., 2015; Shi et al., 2015). Compared to river sand, sea sand slightly reduced the curve curvature of GRACSS at the elastic-plastic stage. This phenomenon is attributed to the sea sand chloride ions increase concrete density whilst improve its porosity and micro-structure (Cheng et al., 2017; Etxeberria et al., 2016). The denser the structure of concrete, the less the plastic deformation of RACSS, and the less the curve curvature of specimen. However, RCA could increase the curve curvature of GRACSS due to the initial defects and older mortar of aggregates (Xiao et al., 2012). It was also observed that the curve slope (in the hardening stage) of specimen simultaneously adopting RCA and sea sand was decreased compared to specimen using river sand and natural coarse aggregates. This is attributed to the properties of RACSS. Huang et al. (2018) reported that the plastic deformation of RACSS was small compared to ordinary concrete. It caused the confinement between core concrete and outer tube was relatively small. The less the confinement, the more the axial deformation of concrete (at the same load level), and the lower the curve slope (hardening stage) of GRACSS (Xiao et al., 2012).

Finally, it can be concluded that the curve curvature (elastic-plastic stage) of GRACSS increased while the curve slope (hardening stage) slightly decreased compared to GCOC. However, due to the high content of sea sand Cl^–, the curve curvature of GRACSS was lower than that of GFRP tube confined recycled aggregate concrete (GRAC).

Strength

The strength (f_cc) of GRACSS is listed in the Table 3 (f_cc= N_max/(A_c)). It can be obtained that type of fine aggregate and RCA replacement percentage had obvious effects on the f_cc of GRACSS.

Sea sand and RCA could reduce the f_cc of specimen. It was found that f_cc of GRACSS-0-R was 6.98% higher than that of GRACSS-100-S. That can be attributed to RCA contain many initial defects and the high sea sand Cl^– content (Li et al., 2017). Therefore, f_cc of GFRP tube confined recycled aggregate concrete with sea sand was low compared with GCOC. It was also found that the influence of sea sand on f_cc was relatively small compared to RCA. The f_cc of GRACSS-0-S was about 1.65% higher than that of GRACSS-100-R. This is mainly because of sea sand Cl^– improve the micro-structure of materials (Xiao et al., 2018).

Generally, the more the contents of RCA and Cl^–, the lower the strength of GRACSS. However, compared to GCOC, the strength reduction of GRACSS was less than 10% due to the high confinement provided by FRP tube.

Peak strain

The peak strain (ε_cc) of GRACSS is defined as the ratio of the peak deformation (△_cc) to the middle span of specimen (L = 400 mm). Details of ε_cc are listed in Table 3. It can be found that RCA replacement percentage and type of fine aggregates changed ε_cc of specimen.

Test results showed that sea sand decreased ε_cc of GRACSS compared to river sand, while RCA could offset the negative of sea sand on ε_cc. It was shown that ε_cc of GRACSS-100-R was 6.91% and 13.2% higher than that of GRACSS-0-R and GRACSS-100-S, respectively. This phenomenon can be attributed to the coupled effects of sea sand and RCA on the micro-and macro-structures of concrete (Huang et al., 2018). Generally, ε_cc of GRACSS was averagely 5.59% lower than that of GCOC.

Transverse deformation development

The transverse deformation development of GRACSS is shown in Figure 6. It can reflect the confinement between outer tube and core concrete. The transverse strain of specimen increased slowly at the early stage of loading. The confinement between outer tube and core concrete was negligible. However, the transverse strain developed rapidly when axial stress (σ) approached 0.9 to 1.0 f_c. The composite action of specimen became obvious. It is attributed to the transverse deformation of core concrete grows quickly compared to GFRP tube when concrete is in the elastic-plastic stage (Li et al., 2015). Furthermore, the confinement between recycled aggregate concrete and GFRP tube (GRACSS-100-R) was high compared to concrete adopting sea sand and gravel (GRACSS-0-R, GRACSS-0-100). It was also found that sea sand obviously decreased the positive influence of RCA on the transverse strain development. Compared to GRACSS-100-R, GRACSS-100-S transverse strain developed slowly at the same load level.

Figure 6.

The development of transverse strain.

Finally, it can be obtained that the transverse deformation development of these specimens increased by the order of GRACSS, GCOC, and GRAC based on the above analyses.

Strength index

The strength index (μ) was used to study the effects of fine and coarse aggregates on the enhancement of GRACSS strength. Generally, μ can be calculated as (equation (1))

μ = N_{max} / f_{c} A_{c},

(1)

where f_c, A_c are axial compressive strength and cross-sectional area of concrete, respectively. $N_{max}$ denotes the peak load.

Details of μ are listed in Table 3. It was obtained that μ of GRACSS was more than 2.69 (2.69–2.86), which shows a good composite action of specimen. The μ slightly changed with the variations of both the RCA replacement percentage and the type of fine aggregate. It was found that μ of GFRP tube confined ordinary concrete (GRACSS-0-R) was 3.74% higher than that of GRACSS-100-S. This can be attributed to the plastic deformation and the cracking of RACSS were lower than those of ordinary concrete (Huang et al., 2018). The more the plastic deformation and the cracking of concrete, the higher the confinement between outer tube and core concrete, and the more the strength index of specimen.

Sea sand reduced the value of μ due to the denser structure of concrete. Test results indicated that μ of GRACSS-0-S was 1.91% lower than that of GRACSS-0-R. However, the influence of RCA on μ was positive. It was found that μ of GRACSS-100-R was 1.47% higher than that of GRACSS-0-R. This is because RCA contain many initial defects and older mortar (Etxeberria et al., 2016). It caused the plastic deformation of recycled aggregate concrete was high compared to concrete adopting natural coarse aggregates.

Finally, it can be concluded that the strength index of GRACSS was similar to that of GCOC. The composite action between GFRP tube and RASCC was acceptable.

Therotical analysis

Peak load

An analytical expression for the peak load is necessary for the design and analysis of GRACSS structure. The specific expression can be obtained based on the test results and related studies (Lam and Teng, 2004; Li et al., 2018). The expression was derived through the following procedures.

(1) Based on Lam and Teng (2004), the peak load (N_max) of GFRP tube confined ordinary concrete (GCOC) was related to the properties and size of GFRP tube and core concrete. The formula of N_max was expressed as (equation (2)):

N_{max} = A_{c} f_{c} (1 + k \frac{f_{l}}{f_{c}}) = A_{c} f_{c} (1 + k \frac{f_{f} A_{f}}{f_{c} A_{c}}),

(2)

where A_c and A_f are cross-sectional areas of core concrete and outer tube, respectively; f_l is the confining pressure provided by GFRP tube, f_f represents transverse tensile strength of tube, k is the parameter.

(2) It was also found that the effects of Cl^– content ( $φ$ ) and recycled coarse aggregates replacement percentage ( $γ$ ) on N_max of confined concrete members was obvious (Huang et al., 2020; Li et al., 2018). The expression of N_max of GRACSS should take the $γ$ and $φ$ into account (equation (3)).

N_{max} = f (φ, γ) A_{c} f_{c} (1 + k \frac{f_{f} A_{f}}{f_{c} A_{c}})

(3)

In equation (3), $f (φ, γ)$ is the expression of influences of $φ$ and $γ$ on the peak load.

(3) The parameter k and the expression of $f (φ, γ)$ can be obtained based on the test results and the related studies (Lam and Teng, 2004; Li et al., 2018). Using the least square method to derive k and $f (φ, γ)$ , and the Matlab software was used to code the related program for the least square method. The peak load of GRACSS is derived according to the above procedures.

N_{max} = (1 - 0.0237 γ - 0.1148 φ) A_{c} f_{c} (1 + 1.76 ξ)

(4)

In equation (4), $A_{c}$ and $f_{c}$ denote cross-sectional area and axial compressive strength of concrete, respectively. $γ$ is the RCA replacement percentage, $φ$ represents sea sand Cl^– content. $ξ$ ( $ξ = \frac{f_{f} A_{f}}{f_{c} A_{c}}$ ) represents the constraining factor.

The comparison between the calculated peak load and the real ones is listed in Table 3. It can be found that the difference was small. It should also be noted that equation (4) is applicable to GRACSS with height-to-diameter ratio of 2.5-5.0.

Stress-strain curve of GRACSS

The mechanical behaviors of GRACSS changed with the variations of both the RCA replacement percentage and the type of fine aggregates. A specific stress-strain expression (equation (5)) for GRACSS under the axial compression is obtained based on the test results and related analyses (Li et al., 2015; Wu and Jiang, 2013; Xiao et al., 2010). It can describe the variation of mechanical properties of specimen. The derivation of expression was given through the following procedures.

(1) The shape of curve of GRACSS was similar to that of GCOC according to the test results and related studies (Li et al., 2015; Wu and Jiang, 2013; Wu and Lv, 2003). Therefore, the stress-strain expression for GRACSS under the axial compression can refer to that for FRP confined ordinary concrete. The specific expression was given (equation (5)):

{\begin{matrix} σ = E_{c} ε + \frac{{(E_{c} - E_{2})}^{2}}{4 f_{0}} ε^{2} (0 \leq ε < ε_{t}) \\ σ = f_{0} + E_{2} ε (ε_{t} \leq ε \leq ε_{cc}) \end{matrix}

(5a, b)

where $σ$ and $ε$ are axial stress ( $σ = N / A_{c}$ ) and strain, respectively. $E_{c}$ is the elastic modulus of core concrete.

(2) However, the effects of sea sand Cl^– content and recycled coarse aggregates replacement percentage were not considered in equation (5). The calculated stress-strain curves were slightly different from the test results. Therefore, the parameters of equation (5) (E₂ and $ε_{cc}$ , etc. ) should be modified according to the test results and studies (Li et al., 2015; Wu and Jiang, 2013). The least square method was adopted to obtain these parameters in this study (the Matlab software was used to code the related program).

The parameters $E_{2}$ and $f_{0}$ can be obtained according to equations (6) and (7).

E_{2} = \frac{f_{cc} - f_{0}}{ε_{cc}}

(6)

f_{0} = (1 - 0.0115 γ + 0.132 φ) {(\frac{f_{l}}{f_{c}})}^{- 0.57} f_{c}

(7)

In equations (6) and (7), $f_{cc}$ and $ε_{cc}$ are the peak stress and strain of GRACSS, respectively. $f_{l}$ is the confining pressure provided by outer tube. According to the test results and analyses, $f_{cc}$ , $ε_{cc}$ , and $f_{l}$ are expressed as equations (8)–(10):

f_{cc} = \frac{N_{max}}{A_{c}}

(8)

f_{l} = \frac{2 f_{f} t}{D}

(9)

ε_{cc} = (2 + 9.6 ξ) ε_{c},

(10)

where $ε_{c}$ denotes the peak strain of plain concrete, t and D are thickness and diameter of outer tube, respectively.

A comparison between calculated curves and test results is shown in Figure 7 and the difference is acceptable. The suggested stress-strain relation can be directly used in the design and analysis of GRACSS structures.

Figure 7.

A comparison between the calculated results and test results: (a) GRACSS-0-R and (b) GRACSS-100-S.

Simulation analysis

In order to study the mechanical behaviors of GRACSS under different conditions, the nonlinear finite element analyses have been undertaken by using ANSYS (ANSYS, 2012).

Type of finite element

Concrete element

Three-dimensional solid element solid65 was selected to model the core concrete. Solid65 is capable of plastic deformation and crushing in compression, cracking in tension. The most advantage of this element is the treatment of nonlinear properties of material.

Furthermore, the parameters of finite element solid65 were important for calculating the nonlinear properties of concrete. Based on the related studies (Xiao et al., 2014), the failure criterion of Willam-Warnke was adopted, and the multi-linear isotropic hardening rule was used to describe the changing of the yield surface with progressive yielding. Finally, in order to model the stress transmitted the closed or opening cracks; the shear transfer coefficients β_c and β_t were 0.25 and 0.94, respectively.

FRP element

The solid46 (three-dimensional solid element) was used to model the outer GFRP tube. Solid46 has 8 nodes and is layered structural element. Compared to shell element, it takes the transverse shear stiffness at the top and bottom surface of the element into account. Solid46 can also well reflect the anisotropic properties of GFRP tube.

Furthermore, the Rankine maximum stress criterion was used, and the liner elastic constitutive model of solid46 was adopted due to the brittle properties of GFRP tube (ANSYS, 2012).

Mechanical properties of materials

The theoretical stress-strain relationship of recycled coarse aggregate concrete with sea sand (RACSS) is given by

y = {\begin{matrix} ax + (3 - 2 a) x^{2} + (a - 2) x^{2} 0 \leq x < 1 \\ \frac{x}{b (x - 1)^{2} + x} x \geq 1, \end{matrix}

(11)

where $y = \frac{σ_{a}}{f_{c}}$ and $x = \frac{ε_{a}}{ε_{c}}$ (equation (11)). $f_{c}$ and $ε_{c}$ are axial compressive strength and peak strain of plain concrete. $σ_{a}$ and $ε_{a}$ represent stress and strain. In equation (11), a and b are the parameter in the stress-strain relationship (Huang et al., 2018). These parameters are expressed as equations (12) and (13):

\begin{matrix} a = 2.27 [(0.755 + 4.3 \times 10^{5} φ^{2}) r^{2} \\ - (1.23 - 1.6 φ) r + 1.061] \end{matrix}

(12)

b = 2.23 [(0.1435 - 245 φ) r + 1.2377]

(13)

where $γ$ denotes RCA replacement percentage, and $φ$ is Cl^– content in sea sand.

Furthermore, the elastic model is adopted for GFRP tube due to it is a kind of brittle materials (Li et al., 2015).

Comparison of finite element analyses and test results

The finite element model used in the analyses is shown in Figure 8. Based on the model, a comparison between test results and calculated results is displayed in Figure 9. It can be found that the test results were well closed to the simulation ones. Therefore, extensive finite element simulations were undertaken to study the mechanical behavior of GRACSS. The typical influences of concrete strength, tube thickness, sea sand Cl^– content and RCA replacement percentage were investigated.

Figure 8.

The illustration of FEM model.

Figure 9.

The comparison between test results and finite element results: (a) GRACSS-0-R and (b) GRACSS-100-S.

Analysis of simulation results

Tube thickness and concrete strength

To study the mechanical behavior of GRACSS under different tube thicknesses and concrete strength, the simulation analyses have been conducted. Details of simulation results are obtained as follows.

Tube thickness

Typical influence of GFRP tube thickness on the mechanical behavior of GRACSS is shown in Figure 10. The peak load and strain obviously increased with an increase in tube thickness. This phenomenon is attributed to the composite action of specimen (Xiao et al., 2010). The thicker the outer tube, the higher the composition action of GRACSS, and the larger the peak load and strain. Furthermore, supportive simulation results indicated that the curve curvature (in elastic-plastic stage) decreased while the curve slope (hardening stage) increased with an increase in tube thickness due the confinement between outer tube and core concrete.

Figure 10.

Mechanical behaviors of GRACSS under different conditions: (a) tube thickness and (b) concrete strength.

Concrete strength

Through the studies of specimen with core concrete strength of C30, C40 and C60, it can be obtained that the peak load of GRACSS increased with an increase in concrete strength (Figure 10). The peak load of GRACSS with C60 was higher than that of specimen adopting C30. However, the influence of concrete strength on the peak strain was relatively small. It was found that the peak stain of specimen with C60 was somewhat lower than that of specimen adopting C40. This may be due to the plastic deformation and cracking of concrete decrease with the increasing strength (Etxeberria et al., 2016). In addition, the concrete strength decreased the curve curvature of GRACSS.

RCA replacement percentage

The axial load-strain curves of GRACSS under different RCA replacement percentages (γ) are shown in Figure 11. It can be obtained that the peak load (N_max) of specimen decreased while the peak strain (ε_cc) increased with an increase in γ. The N_max (ε_cc) of GRACSS-0-0.1488 (Specimen was denoted by “GRACSS-RCA replacement percentage-sea sand Cl^– content) was 2.38%, 3.73%, and 5.56% (2.82%, 6.34%, and 10.8%) higher (lower) than those of GRACSS-30-0.1488, GRACSS-70-0.1488, and GRACSS-100-0.1488, respectively. Furthermore, the positive influence of RCA on ε_cc decreased with the increasing Cl^– content (φ). The peak strains of GRACSS-0-0, GRACSS-30-0, GRACSS-70-0, and GRACSS-100-0 was 20.7%, 21.9%, 16.8%, and 13.9% more than those of GRACSS-0-0.1488, GRACSS-30-0.1488, GRACSS-70-0.1488, and GRACSS-100-0.1488, respectively.

Figure 11.

Axial load-strain curve of GRACSS under different RCA replacement percentage: (a) 0% sea sand Cl^– content and (b) 0.1488% sea sand Cl^– content.

The shape of GRACSS curve changed with the variations of γ. Generally, the more the γ, the lower the initial stiffness, and the higher the curve curvature (elastic-plastic stage). That can be attributed to many initial defects and older mortar in RCA.

Sea sand Cl^– content

Typical effects of sea sand Cl^– content on the axial load-strain curves of GRACSS are shown in Figure 12. The test results indicated that N_max of GRACSS slightly changed with increasing φ. Compared with GRACSS-100-0.255 (The specimen was denoted by “GRACSS-RCA replacement percentage-sea sand Cl^– content), the variations of N_max of GRACSS-100-0 and GRACSS-100-0.1488 were relatively small. However, ε_cc decreased with an increase in φ. The peak strain of GRACSS-100-0 was 13.9% and 27.3% more than that of GRACSS-100-0.1488 and GRACSS-100-0.255, respectively.

Figure 12.

Axial load-strain curve of GRACSS under different sea sand Cl^– contents.

Furthermore, Sea sand changed the shape of axial load-deformation curve of GRACSS under compression. The initial stiffness of curve increased while the curve curvature decreased with increasing Cl^– content (Figure 12). This phenomenon can be explain as the sea sand chloride ions increase concrete density whilst improve its porosity and micro-structure (Cheng et al., 2017; Etxeberria et al., 2016). The denser the micro-structure of concrete, the less the plastic deformation of specimen, and the less the curve curvature of specimen.

Conclusion

The mechanical behaviors of GFRP tube confined recycled aggregate concrete with sea sand (GRACSS) under the axial compression were tested and analyzed in this paper. The following conclusions can be drawn.

Typical failure patterns of GFRP tube confined recycled aggregate concrete with sea sand are similar to those of GFRP tube confined ordinary concrete (GCOC). The axial load-deformation curve of GRACSS consists of elastic-plastic stage and hardening stage. There is also a transition part between the two stages. The curve curvature and curve slope of GRACSS are different from those of GFRP tube confined ordinary concrete. Sea sand and recycled coarse aggregates (RCA) slightly influence the shape of the load-deformation curve.

The peak load of GRACSS is averagely 7.04% lower than that of GCOC. Sea sand and RCA reduce the strength and deformation of specimens.

The strength factor and transverse deformation development of GRACSS are varied with both the RCA replacement percentage and the type of fine aggregates. The strength factor of GRACSS is 3.74% lower than that of GCOC. Sea sand delays the development of transverse deformation, while RCA could improve the transverse deformation.

An analytical expression for the peak load of GRACSS is suggested. Based on the experimental data, a numerical stress-strain relation of GRACSS considering the influences of sea sand Cl^– content and RCA replacement percentage is also proposed.

The peak load and strain of GRACSS increase with an increase in the tube thickness according to the finite element analysis. The peak deformation of specimens is varied with the concrete strength. The difference between the peak load of GCOC and that of GRACSS is less than 10%. However, the GRACSS peak deformation obviously decreases with the increasing RCA and Cl^– contents. The load-deformation curve of GRACSS changes with the variations of tube thickness, concrete strength, RCA replacement percentage and sand Cl^– content.

Footnotes

Appendix

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 study is supported by the National Natural Science Foundation of China (No. 51408346, No. 51978389), the Systematic Project of Guangxi Key Laboratory of Disaster Prevention and Structural Safety (2019ZDK035) and the Opening Foundation of Shandong Key Laboratory of Civil Engineering Disaster Prevention and Mitigation (No. CDPM2019KF12).

ORCID iD

Jianzhuang Xiao

References

ANSYS (2012) ANSYS 14.0 help system-release note. Canonsburg: ANSYS Inc.

Bravo

Brito

Pontes

, et al. (2015) Mechanical performance of concrete made with aggregates from construction and demolition waste recycling plants. Journal of Cleaner Production 99: 59–74.

Cheng

Shui

Sun

, et al. (2017) Effects of fly ash, blast furnace slag and metakaolin on mechanical properties and durability of coral sand concrete. Applied Clay Science 141: 111–117.

Choi

Kim

, et al. (2014) Effect of corrosion method of the reinforcing bar on bond characteristics in reinforced concrete specimens. Construction and Building Materials 54: 180–189.

Dias

WPS

Seneviratne

GAPSN

Nanayakkara

SMA

(2008) Offshore sand for reinforced concrete. Construction and Building Materials 22(7): 1377–1384.

Dolage

(2013) Offshore sand as a fine aggregate for concrete production. British Journal of Applied Science and Technology 3(4): 813–825.

Etxeberria

Gonzalez-Corominas

Pardo

(2016) Secondary aggregates and seawater employment for sustainable concrete dyke blocks production: Case study. Construction and Building Materials 113: 586–595.

Geng

Wang

Chen

(2015) Time-dependent behavior of recycled aggregate concrete-filled steel tubular column. Journal of Structural Engineering-ASCE 141(10): 146–158.

Huang

Wang

, et al. (2018) Mechanical properties of sea sand recycled aggregate concrete under axial compression. Construction and Building Materials 175: 55–63.

10.

Huang

Wang

(2020) Behaviour of sea sand recycled concrete filled steel tube under axial compression. PICE-Structures and Buildings 173(4): 302–312.

11.

Kato

Yamaguchi

(1997) Computational model for deterioration of aramid fiber by ultraviolet rays. In: Proceedings of the third international symposium non-metallic (FRP) reinforcement for concrete structures, Sapporo: Japan Concrete Institute, pp.163–170.

12.

Kourmpanis

Papadopoulos

Moustakas

, et al. (2008) Preliminary study for the management of construction and demolition waste. Waste Management Research 26(3): 265–274.

13.

Kumar

BNN

Kumar

Babu

, et al. (2016) An experimental study on sea sand by partial replacement of sea sand in concrete. International Journal of Scientific Research in Science and Technology 2(2): 181–184.

14.

Lam

Teng

(2004) Ultimate condition of fiber reinforced polymer-confined concrete. Journal of Composites for Construction 8(6): 539–548.

15.

Wang

(2013) Durability of FRP rebars in alkaline environment of concrete. Building Science 29(1): 47–51. (in Chinese)

16.

Geng

Huang

, et al. (2015) Chloride resistance of concrete with metakaolin addition and seawater mixing: A comparative study. Construction and Building Materials 101: 184–192.

17.

Long

Tam

VWY

, et al. (2017) Effects of nano-particles on failure process and microstructural properties of recycled aggregate concrete. Construction and Building Materials 142: 42–50.

18.

Zhao

Raman Singh

, et al. (2016a) Experimental study on seawater and sea sand concrete filled GFRP and stainless steel tubular stub columns. Thin-Walled Structures 106: 390–406.

19.

Zhao

Raman Singh

, et al. (2016b) Tests on seawater and sea sand concrete filled CFRP, BFRP and stainless steel tubular stub columns. Thin-Walled Structures 108: 163–184.

20.

Teng

Zhao

, et al. (2018) Theoretical model for seawater and sea sand concrete-filled circular FRP tubular stub columns under axial compression. Engineering Structure 160: 71–84.

21.

Limeira

Etxeberria

Agullo

, et al. (2011) Mechanical and durability properties of concrete made with dredged sea sand. Construction and Building Materials 25(11): 4165–4174.

22.

Ministry of Communications of the People’s Republic of China (MCPRC) (2010) GB/T 25177-2010: Recycled Coarse Aggregate for Concrete. Beijing: MCPRC.

23.

MCPRC (2011a) GB/T 14685-2011: Pebble and Crushed Stone for Construction. Beijing: MCPRC.

24.

MCPRC (2011b) GB/T 14684-2011: Sand for Construction. Beijing: MCPRC.

25.

Poon

Kou

Wan

, et al. (2009) Properties of concrete blocks prepared with low grade recycled aggregates. Waste Management 29(8): 2369–2377.

26.

Shi

Shui

, et al. (2015) Combined effect of metakaolin and sea water on performance and microstructures of concrete. Construction and Building Materials 74(15): 57–64.

27.

Teng

Xiang

, et al. (2019) Development and mechanical behaviour of ultra-high-performance seawater sea-sand concrete. Advances in Structural Engineering 22(14): 3100–3120.

28.

(2003) Study on the stress-strain relationship of FRP-confined concrete circular column without a strain-softening response. Journal of Building Structures 24(5): 1–11. (in Chinese)

29.

Jiang

(2013) Effective strain of FRP for confined circular concrete columns. Composite Structures 95: 479–491.

30.

Xiao

Huang

Sun

(2014) Seismic behavior of recycled aggregate concrete filled steel and glass fiber reinforced plastic tube columns. Advances in Structural Engineering 17(5): 693–705.

31.

Xiao

Huang

Yang

, et al. (2012) Mechanical properties of confined recycled aggregate concrete under axial compression. Construction and Building Materials 26(1): 591–603.

32.

Xiao

Zhang

, et al. (2018) Basic mechanical properties of seawater sea-sand recycled concrete. Journal of Architecture and Civil Engineering 35(2): 16–22. (in Chinese)

33.

Xiao

Teng

(2010) Behavior and modeling of confined high-strength concrete. Journal of Composites for Construction 14(3): 249–259.

34.

Zhang

Xiao

Liao

, et al. (2019) Structural behavior of seawater sea-sand concrete shear wall reinforced with GFRP bars. Engineering Structure 189: 458–470.

Mechanical behaviors of GFRP tube confined recycled aggregate concrete with sea sand

Abstract

Keywords

Introduction

Experimental program

Raw materials

Specimen design

Test setup, instrumentation, and loading protocol

Experimental results

Failure process

Axial load-deformation curve

Strength

Peak strain

Transverse deformation development

Strength index

Therotical analysis

Peak load

Stress-strain curve of GRACSS

Simulation analysis

Type of finite element

Concrete element

FRP element

Mechanical properties of materials

Comparison of finite element analyses and test results

Analysis of simulation results

Tube thickness and concrete strength

Tube thickness

Concrete strength

RCA replacement percentage

Sea sand Cl– content

Conclusion

Footnotes

Appendix

Declaration of conflicting interests

Funding

ORCID iD

References

Sea sand Cl^– content