Mechanical behavior of recycled aggregate concrete-filled steel tube stub columns after exposure to elevated temperatures

Construction and Building Materials 146 (2017) 571–581

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Mechanical behavior of recycled aggregate concrete-filled steel tube stub columns after exposure to elevated temperatures Wengui Li a,⇑, Zhiyu Luo a,b, Zhong Tao c, Wen Hui Duan d, Surendra P. Shah e a

Centre for Built Infrastructure Research, School of Civil and Environmental Engineering, University of Technology Sydney, Sydney, NSW 2007, Australia College of Civil Engineering, Hunan University, Changsha, Hunan 410082, China c Centre for Infrastructure Engineering, Western Sydney University, Penrith, NSW 2751, Australia d Department of Civil Engineering, Monash University, Clayton, VIC 3800, Australia e Center for Advanced Cement-Based Materials (ACBM), Northwestern University, Evanston, IL 60208, USA b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


The heated RACFST stub columns

manifested local outward folding and buckling.
The ductility of RACFST stub columns increased with the increase in high temperature.
The compressive strength declined dramatically heated by high temperatures of 500 °C and 700 °C.
The elastic modulus of RACFST decreased linearly with the increase in high temperature.
The RACFST stub columns showed higher peak strain than that of the corresponding NACFST column.

a r t i c l e

i n f o

Article history: Received 9 January 2017 Received in revised form 8 March 2017 Accepted 14 April 2017

Keywords: Recycled coarse aggregate (RCA) Recycled aggregate concrete-filled steel tubes (RACFST) Elevated temperature Compression Mechanical behavior

⇑ Corresponding author. E-mail address: [email protected] (W. Li). http://dx.doi.org/10.1016/j.conbuildmat.2017.04.118 0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.

a b s t r a c t The compressive mechanical behaviors of recycled aggregate concrete-filled steel tube (RACFST) stub columns after exposure to elevated temperatures were experimentally investigated in this study. The RACFST stub columns incorporating different recycled coarse aggregate (RCA) replacement ratios of 0, 50% and 100% were heated under elevated temperatures of 200 °C, 500 °C, and 700 °C. The results show that the compressive strength and elastic modulus of RACFST columns were relatively inferior to those of the corresponding natural aggregate concrete-filled steel tube (NACFST) columns after exposure to the same elevated temperatures, and the degradations became more pronounced with increasing RCA replacement ratio and higher temperature. This phenomenon might be attributed to the lower resistance of recycled aggregate concrete (RAC) than natural aggregate concrete (NAC) when was exposed to elevated temperatures. However, after elevated temperature exposure, the peak strain of RACFST stub column was relatively higher than that of the NACFST counterpart. Degradation regression formulas of mechanical properties and deformation behaviors of RACFST stub columns after exposure to elevated temperatures were proposed and agreed well with the experimental results. Ó 2017 Elsevier Ltd. All rights reserved.

572

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

Nomenclature Ac As Es(0) Es(r) Es(20) Es(T) fy fck Ke Ke,c Ke,e RCAKe RCAKe Ke Ker,c Ke,e Nu(0)

cross-sectional area of concrete, mm2 cross-sectional area of steel, mm2 elastic modulus of the NACFST stub column, N/mm2 elastic modulus of RACFST stub column with different RCA replacement ratio, N/mm2 elastic modulus of specimens at room temperature of 20 °C, N/mm2 elastic modulus of specimens after exposed to temperature T, N/mm2 yield strength of steel, N/mm2 characteristic concrete compressive strength (0.67 fcu for normal strength of concrete, N/mm2 residual elastic modulus ratio calculated residual elastic modulus ratio experimental residual elastic modulus ratio RCA elastic modulus ratio RCA residual peak strain ratio residual peak strain ratio calculated residual peak strain ratio experimental residual elastic modulus ratio load capacity of NACFST, kN

1. Introduction After demolition of old concrete buildings and the occurrence of natural disasters like earthquakes and hurricanes, much solid waste concrete is generated, usually associated with serious environmental problems. Reprocessing the waste concrete into recycled aggregates is an ideal way of solving the environmental problem and can also conserve natural aggregate resources. However, the promotion of recycled aggregate concrete (RAC) is restricted by its relatively poorer mechanical properties than those of natural aggregate concrete (NAC) [1,2]. To promote the usage of recycled coarse aggregate (RCA) and improve the mechanical strength and stiffness of RAC, Konno et al. [3] recommended confining RAC within steel tubes. Then the concept of rAC-filled steel tubes (RACFST) was developed by subsequent researchers. The mechanical performances of RACFST have been explored in several studies [4,5]. Results have showed that, due to the confinement effect of steel tube, compared with corresponding plain RAC, the axial compressive strength and deformation of the RACFST were significantly improved. Studies of uniaxial mechanical behavior, creep, and flexural strength have indicated that the performance of RACFST is inferior to that of the corresponding NACfilled steel tubular (NACFST) columns [6–9], but RACFST exhibits a really efficient way of reducing the adverse impact on the natural environment caused by the defects of RCA, and it is feasible and safe to use RACFST columns in structural engineering applications. Up to now, few studies have investigated the mechanical behaviors of RACFST columns after exposure to fire or elevated temperatures. Fire is a frequently occurring disaster, and is one of the most serious threats to buildings. Understanding the fire resistance of RACFST is essential for its safe usage in building structures. Studies of the fire resistance of NACFST columns have been conducted by many researchers. For example, Han et al. [10,11] investigated the compressive and flexural behavior of NAC-filled steel tubes after exposure to iSO-834 standard fire [12]. Abbas et al. [13] experimentally studied the effect of cooling regimes on performance of fire damaged NACFST columns. Jiang et al. [14] explored the mechanical behavior of post-fire NACFST column subjected to

Nu(r) Nu(20) Nu(T) RCARSI RSI RSIc RSIrc RSIe T r e(0) e(r)

e(20) e(T) n

load capacity of RACFST stub column with different RCA replacement ratio r, kN compressive strength of NACFST stub column at room temperatures of 20 °C, kN strength of NACFST stub column after exposed to high temperature T, kN RCA residual strength index residual strength index calculated residual strength index calculated residual strength index by modified formula experiment residual strength index high temperature, °C RCA replacement ratio peak strain of the NACFST stub column peak strain of RACFST stub column with RCA replacement ratio r peak strain of RACFST at room temperature of 20 °C peak strain of RACFST after exposed to high temperature T confinement factor

biaxial force and bending. Dinh et al. [15] proposed a finite element analysis model for studying the effects of steel type and crosssection shape on the fire resistance of NACFST columns. Using a three-dimensional finite element model, Yao et al. [16] analyzed the performance of fire-exposed NACFST stub columns under different heating conditions. Moreover, some design standards exist to guide practical design for the fire resistance of NACFST columns [17,18]. However, limited studies on the fire resistance of RACFST columns have been reported. Only Yang et al. [19] investigated the mechanical behavior of RACFSTs after exposure to elevated temperatures of 300 °C, 600 °C and 800 °C. Formulas evaluating the strength ratio and elastic modulus ratio for NACFSTs were used for RACFST specimens. It is concluded that the predictions of residual strength and residual elastic modulus of RACFST from these formulas used for fire-damaged NACFST columns before are lower than experimental results. In contrast to the extensive research into NACFST columns, the limited existing research is insufficient to reflect the fire resistance of RACFST columns. Furthermore, more accurate formulas are needed and higher temperature ranges should be adopted. In this study, the compressive mechanical behaviors of RACFST stub columns were investigated after exposure to high temperatures of 200 °C, 500 °C and 700 °C. The effect of RCA replacement ratios of 0, 50%, and 100% on the mechanical behaviors of the RACFST stub columns after exposure to elevated temperatures was also analyzed. Degradation regression formulas are proposed for the mechanical behaviors of RACFST stub columns including compressive strength, elastic modulus, and peak strain after exposure to elevated temperatures. The values calculated by the modified formulas agreed well with the experimental results. The associated findings can provide useful insights into the assessment and retrofitting of fire-damaged RACFST structures. The experimental work also can provide a very good basis for conducting future theoretical analysis on the post-fire behavior of RACFST columns.

573

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

temperatures, one RACFST specimen was prepared with high quality for different testing condition in this study.

2. Experimental program 2.1. Material properties

2.3. Testing methods Ordinary Portland cement of grade 42.5, river sand as fine aggregate and tap water were used in specimen preparation. Gravels from an aggregate production plant in Changsha, China were used in this test as NCA. NCA had water absorption of 0.67% and bulk density of 1420 kg/m3. RCA was obtained from abandoned ground concrete which was crushed by a jaw crusher. The RCA had a water absorption is 8.54%, and bulk density of 1235 kg/m3. To meet the requirements of the Chinese code GB50204-2002 [20] and ACI Committee 318 [21], the grading of coarse aggregate in this test was 5–20 mm, so that the maximum aggregate size did not exceed a quarter of the minimum dimension of the structural component’s cross-section. The physical properties and size grading of NCA and RCA are shown in Tables 1 and 2. The mix design proportion of RAC was made by mass because of the content variation of old cement mortar adhered to the original aggregate. The effective water to cement ratio of RAC in the mixing design was set as 0.5. Seamless steel tubes (Q235) with the outer diameter of 96 mm and wall thickness of 2.0 mm were purchased from a local steel market in Changsha, China. Three duplicate steel samples cut from steel tubes were used to measure the mechanical properties of the steel. The measured average yield strength and modulus of elasticity of the steel were 320.1 MPa and 201.5 GPa, respectively. The RCA replacement ratios for RAC mix design were 0, 50% and 100%, and the RCA replacement ratio referred to the ratio of the mass of RCA to the total mass of coarse aggregates used for the RAC mix design. Thus, the NCA was replaced with RCA by weight. Owing to the high water absorption of RCA, apart from the mixing water, additional water was added for the RAC to achieve the same effective water to cement ratio, as determined by the amount of water required to change the RCA from its natural state to a saturated surface dry condition [22,23]. The mix proportions of these three kinds of concrete are presented in Table 3. After exposure to elevated temperatures, the compressive strengths of RAC cubes with 0, 50%, and 100% RCA replacement ratios were 42.8 MPa, 43.66 MPa and 35.27 MPa, respectively.

2.2. Specimen preparation The steel tubes were cut into lengths of 600 mm. A small wooden plate was pasted onto each end of steel tubes to prevent mortar leakage. Then RAC was poured into each steel tube layer by layer, and vibrated by a poker vibrator. The RACFST specimens were placed upright to air-dry in the laboratory, after which they were cut to a height of 288 mm by a mechanical processor. The length to diameter ratio was 3.0 in order to minimize the effects of overall buckling and end conditions, and this ratio also has been widely applied in other experiments [19,24]. The manner of cutting by a mechanical processor was to make sure the end surfaces of the specimens were sufficiently smooth. A double sided grinding machine was also used to ensure that the two ends of each specimen were precisely parallel. A typical RACFST stub column specimen used for experiments is shown in Fig. 1. For the compression test on RACFST columns after exposure to elevated

Elevated temperature exposure of RACFST stub columns was carried out using electrical furnace at the Key Laboratory of Building Safety and Energy Efficiency (Hunan University), Ministry of Education, Changsha China. The heating rate was set at 10 °C per minute [25]. After each specified maximum temperature was reached, the temperature was maintained for 3 h to ensure that the specimens were heated evenly and achieved a stable residual strength [26]. Then, the RACFST specimen was allowed to cool down until to room temperature. The mechanical behavior of fire-exposed RACFST stub columns was tested by a testing machine with 3000 kN compressive capacity. Four strain gauges glued at the mid-height position of the specimens with an interval of 90° and two displacement transducers in symmetrical positions were used to measure the longitudinal deformation. The experimental setup of the RACFST stub column under compression is shown in Fig. 2. The load increment was set as one tenth of the estimated load capacity and changed to about 5% estimated strength after the steel tube yielded. Each load level was maintained for about 2 min. When a specimen was close to failure, a slow and continuous loading rate was used. During the loading process, load capacity, deformation, and failure patterns of the specimens were regularly captured and recorded. 3. Test results and discussion 3.1. Failure patterns After heated at elevated temperatures, although there were no obvious differences between the failure patterns of different RACFST stub columns, the colors of the outer steel tubes of the specimens clearly differed. At the temperature of 200 °C, the color of the steel tube was slightly darker than that at room temperature of 20 °C. When the temperature was 500 °C, the outer steel tubes became significantly darker than those of both at 20 °C and 200 °C. When the temperature reached 700 °C, the color of the steel tube was brick-red. On the other hand, during the compression process, the brick-red oxide layer on the surface of the steel tube gradually peeled off with the increasing load, and several shear slippage lines at 45° from the horizontal became clearly visible. The shear slippage lines began to appear at around 50–70% of the ultimate load capacity. After the compression test, the failure mode of the outer steel tube was typically a local failure mode (outward folding), which was similar to the failure mode found in previous studies of fire-exposed NACFST stub column [11]. The relevant failure patterns of RACFST stub columns are showed in Fig. 3. It was concluded that the failure pattern of the RACFST stub columns did not change with the increase in the RCA replacement ratio heated at different elevated temperatures. 3.2. Load versus strain curves The compressive load versus strain curves of RACFST stub columns after undergoing different elevated temperatures are shown

Table 1 Physical property of NCA and RCA. Aggregate types

Aggregate grading (mm)

Bulk density (kg/m3)

Apparent density (kg/m3)

Water absorption (%)

NCA RCA

5–20 5–20

1420 1235

2752 2637

0.67 8.54

574

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

Table 2 Size grading of NCA and RCA. Nominal size (mm)

Accumulated retained percentage by square hole screen (mm)

5–20

2.36

4.75

9.50

16.0

19.0

26.5

98

96

54

10

6

0

Table 3 Mix proportions of NAC and RAC. Specimens

Cement (kg/m3)

Sand (kg/m3)

NCA (kg/m3)

RCA (kg/m3)

Mixing water (kg/m3)

Additional water (kg/m3)

NAC RAC50 RAC100

402 402 402

645 645 645

1148 574 0

0 574 1148

205 205 205

0 28 55

Both the effective water to cement ratio of NAC and RAC was 0.5.

(a) Side view

(b) Top view

Fig. 1. Typical RACFST stub column specimen.

(a) Dimensions

ity of the RACFST specimens increased with the increase in temperature. These load-strain curves of RACFST specimens heated to 200 °C were similar to those at 20 °C. Therefore, it can be concluded that there was no obvious impairment of the mechanical properties of the RACFST stub columns at 200 °C. On the other hand, when the temperature increased to 500 °C and 700 °C, the ascending and descending line stages of the load-strain curves became much less distinct, and the peak load capacity also decreased dramatically. The influences of RCA replacement ratios on the load-strain curves of RACFST stub columns after exposure to elevated temperatures are displayed in Fig. 5. It is evident that there are no noticeable differences in the shape of the load-strain curves of RACFST columns with different RCA replacement ratios. The load-strain curve of RAC50FST with RCA replacement ratio of 50% tended to overlap with the curve of the NACFST column without RCA content. This indicates that adding RCA by 50% replacement ratio is unlikely

(b) Test instrumentation

Fig. 2. Experimental setup of RACFST stub column under compression.

in Fig. 4. It was found that the formation of load-strain curves of the RACFST stub columns with different RCA replacement ratios were similar to each other during compression. For instance, with the increase in temperature, the elastic stage of the load-strain curves became shorter and the elastic-plastic stage turned much longer. The compressive load capacity decreased significantly after exposure to higher temperatures, but the deformation and ductil-

to significantly reduce the load capacity of RACFST columns, which provides a positive evident for promoting the application RACFST. This finding is consistent with the compressive strength results of RAC cubes after exposure to elevated temperature. For example, at the temperatures of 20 °C and 200 °C, the RAC50FST stub column even exhibited slightly better capacity than that of the NACFST column without RCA, but at the temperatures of 500 °C and 700 °C,

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

575

(a) NACFST stub columns

(b) RAC50FST stub columns

(c) RAC100FST stub columns Fig. 3. Failure patterns of high temperature treated RACFST stub columns with different RCA replacement ratio.

the load capacity of the RAC50FST stub column was obviously lower than that of the NACFST column. In the RAC100FST stub columns with 100% RCA replacement ratio, the load-strain curves were significantly lower than those of both the NACFST and RAC50FST columns after exposure to elevated temperatures, especially at 700 °C. It means adding 100% RCA obviously decreases the load capacity of RACFST columns after exposure to elevated temperatures. On the other hand, given the same elevated temperature, the load-strain curves of the RACFST stub columns incorporating different RCA replacement ratios almost coincided with each other at beginning of the ascending elastic stage, but this overlap section became progressively shorter with the increase in temperature. With respect to the deformation of RACFST, it should be noted that although the elevated temperatures might induce contraction or expansion of RAC and steel tube, the effect of these residual strains of RAC and steel weren’t considered in the RACFST deformation analysis because of the complex volume changes. The reason is that it is very difficult to determine the actual RAC contraction or steel expansion after exposure to elevated temperatures. 3.3. Compressive strength Yang et al. [19] previously chose the axial load corresponding to the strain of 0.03 as the ultimate strength of RACFST stub columns.

However, it is found that this load level might be not suitable in this study. Peak axial load has been used to represent the load carrying capacity of RACFST stub columns [27]. As shown in Eq. (1), the residual strength index (RSI) has typically been used to evaluate the degradation of strength of NACFST stub columns after exposure to fire or elevated temperature [11,28].

RSI ¼

Nu ðTÞ Nu ð20Þ

ð1Þ

where Nu(T) is defined as the residual strength of a NACFST stub column exposed to a elevated temperature (T), and Nu(20) is the strength of a NACFST stub column at room temperature of 20 °C. In this study, the residual strength index (RSI) of the RACFST stub columns was calculated as shown in Fig. 6. It was found that after exposure to elevated temperatures, RACFST stub columns with all the different RCA replacement ratios showed decreased compressive strength. The ranges of the RSI were 0.969–0.984 at 200 °C, 0.782–0.806 at 500 °C, and 0.623–0.663 at 700 °C, respectively. The decline in compressive strength was not obvious at 200 °C, but a dramatic decline was observed when the temperature was further elevated to 500 °C and 700 °C. In the NAC, the initial degradation of compressive strength usually occurs between 200 °C and 250 °C [29]. The reason is that when the temperature was below 300 °C, the strength of the NAC seems not to decrease significantly and the strength loss could be recovered through

576

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

700 600

Load (kN)

500 400 300 o

NACFST (20 C) o NACFST (200 C) o NACFST (500 C) o NACFST (700 C)

200 100 0 0.000

0.005

0.010

0.015

0.020

0.025

0.030

Strain

(a) NACFST stub columns

tube can also recover most of their original strength and stiffness when it is heated at temperatures below 300 °C [16,31]. As a result, when the temperature reached 200 °C, there was no obvious reduction in the mechanical strength of RACFST stub columns. However, if temperature was above 300 °C, the loss of bound water in the concrete composites became more remarkable, so a prominent loss of strength was usually observed [32]. When the elevated temperature further rose to 600 °C, RCA experienced thermal expansion and the internal stresses caused by elevated temperature gave rise to extensive cracking. Thus, at temperature of 600 °C, the mechanical performance of RAC was severely deteriorated [32]. Moreover, steel tubes could also evince a visible reduction of strength when temperature was above 500 °C [16]. As a result, the compressive strength of RACFST stub columns declined prominently after exposure to temperatures of 500 °C and 700 °C. Like the RSI of RACFST stub columns after exposure to elevated temperatures or fire, the RCA residual strength index (RCARSI) was defined to evaluate the effect of the RCA replacement ratio on the decline in compressive strength of RACFST stub columns damaged by elevated temperatures, and is expressed as Eq. (2).

700

RCARSI ¼ 600

Load (kN)

500 400 300 o

RAC50FST (20 C) o RAC50FST (200 C) o RAC50FST (500 C) o RAC50FST (700 C)

200 100 0 0.000

0.005

0.010

0.015

0.020

0.025

0.030

Strain

(b) RAC50FST stub columns 700 600

Load (kN)

500 400 300 o

RAC100FST (20 C) o RAC100FST (200 C) o RAC100FST (500 C) o RAC100FST (700 C)

200 100 0 0.000

0.005

0.010

0.015

0.020

0.025

0.030

Strain

(c) RAC100FST stub columns Fig. 4. Load-strain curves of RACFST stub columns after exposure to high temperatures.

the rehydration [30]. This conclusion might be also appropriate for the RAC exposed to elevated temperatures. At the same time, steel

Nu ðrÞ Nu ð0Þ

ð2Þ

where Nu(r) is the load capacity of a RACFST stub column with a certain RCA replacement ratio, and Nu(0) refers to the load capacity of a NACFST column without RCA content. The RCARSIs of RACFST stub columns with RCA ratios of 50% and 100% at different temperatures are displayed in Fig. 7. It is clear that the variation in RCARSI is relatively less than that in the RSI, indicating that compared to elevated temperature exposure, the RCA replacement ratio had smaller effects on the compressive strength of RACFST stub columns. In general, there were no significant differences in RCARSIs between RACFST stub columns and the corresponding NACFST columns after exposure to elevated temperatures. For example, the lowest RCARSI was 0.877 for the RAC100FST stub column at temperature of 700 °C. This was due to the increase in the RCA replacement ratio that only slightly reduces the strength of the inner concrete, but the strength of the RACFST stub columns was determined by the composite behavior of both the steel tube and RAC. It was found that as temperature increased, the RCARSI decreased in general (the RCARSI of the NACFST was 1.0), and the RCARSI of the RAC100FST stub column was lower than those of the NACFST and RAC50FST columns. At temperatures of 20 °C and 200 °C, the RCARSI of the RAC50FST stub column was close to 1.0. It indicates that 50% RCA incorporation does not reduce the compressive strength of the RACFST after exposure to the temperature less than 200 °C. But when the temperature reached 500 °C or 700 °C, the RCARSI of the RAC50FST stub column became less than 1.0. These results imply that, along with the increase in elevated temperature, the RACFST stub columns exhibited lower strength than the corresponding NACFST column. The decrease in load-carrying capacity of RACFST was more obvious at high RCA replacement. This effect could be attributed to the weaker fire resistance of RAC compared to NAC, especially at elevated temperatures. It might also be due to the flammable noncementitious impurities attached to the RCA and to the cracks and pores induced in the old mortar matrix after heating [33]. Therefore, it could be attributed to some other negative effects caused by old mortar matrix on the surface of RCA. To be more precise, the coefficients of thermal expansion of old cement mortar, new cement mortar and natural aggregate differed, generating internal stresses and microcracks in both the new and old interfacial transition zones. Therefore, in terms of the microstructures, the RAC had more weak regions compared to the NAC. On the other hand, the old cement mortar in RCA apparently reduced

577

700

700

600

600

500

500

Load (kN)

Load (kN)

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

400 300

400 300

o

NACFST (20 C) o RAC50FST (20 C) o RAC100FST (20 C)

200 100

0.005

0.010

0.015

0.020

0.025

NACFST (200 C) o RAC50FST (200 C) o RAC100FST (200 C)

100 0 0.000

0.030

0.010

0.015

0.020

0.025

Strain

(a) Room temperature of 20 oC

(b) Temperature of 200 oC

700

700

600

600

500

500

400 300 NACFST (500 C) o RAC50FST (500 C) o RAC100FST (500 C)

100

0.005

0.010

0.015

0.020

0.025

300 o

NACFST (700 C) o RAC50FST (700 C) o RAC100FST (700 C)

100

0.030

0.030

400

200

o

200

0 0.000

0.005

Strain

Load (kN)

Load (kN)

0 0.000

o

200

0 0.000

0.005

0.010

0.015

0.020

0.025

0.030

Strain

Strain

(d) Temperature of 700 oC

o

(c) Temperature of 500 C

1.1

1.1

1.0

1.0

0.9

0.9

RCARSI

RSI

Fig. 5. Load-strain curves of high temperature heated RACFST stub columns with different RCA replacement ratios.

0.8

0.7

0.7

NACFST RAC50FST RAC100FST

0.6

0.8

NACFST RAC50FST RAC100FST

0.6 0.5

0.5

20

20

200

500

200

7 00 o

the content of natural aggregate and increased the content of cement mortar in the RAC, which might further affect the fire resistance of the RACFST stub columns. Eq. (3) was proposed by Han [31] to estimate the RSI of NACFST stub column after heating by a high temperature.

700 o

High temperature ( C) Fig. 6. Residual strength index of RACFST stub columns after exposure to different high temperatures.

500

High temperature ( C) Fig. 7. RCARSIs of high temperature treated RACFST stub columns with different RCA replacement ratios.

RSI ¼

1 1 þ 6:5n0:54 ðT  20Þ2:6  108

ð3Þ

where, n is fyAs/fckAc, referring to the confinement factor, in which fy and As are the yield strength and cross-sectional area of steel,

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

respectively; fck and Ac represent the characteristic strength and cross-sectional area of the core concrete, respectively. This formula, Eq. (3), was validated to determine whether it was appropriate for calculating the RSI of RACFST stub columns heated by elevated temperatures. Comparison of calculated RSI values (RSIc) by Eq. (3) and the experimental RSI values (RSIe) are shown in Fig. 8. It indicates that the RSIc are obviously not consistent with the RSIe, especially at elevated temperatures. Thus, RSIc is not suitable for calculating residual strength of the RACFST. For instance, at temperature of 700 °C, the calculated value was just about 60%– 67.9% of the experimental value. Therefore, an adjustment coefficient (g) relevant to the RCA replacement ratio and elevated temperature was applied to modify Eq. (3), on the basis of the experimental results. The modified formulas are given in Eqs. (4) and (5). As shown in Fig. 9, the results obtained by modified RSI formula (RSIrc) agree well with those from the RSIe. Thus, the modified RSI formula could sufficiently estimate the residual strength of RACFST stub columns damaged by elevated temperatures.

RSIr ¼ g

1 1 þ 6:5n

0:54

1.0 T=200 ºC

T=500 ºC

0.6

T=700 ºC

0.4 2

R =0.915

0.2

0.0 0.0

0.2

0.4

0.6

0.8

1.0

RSIe Fig. 9. Comparison between modified formula calculated and experimental residual strength index of RACFST stub columns.

ð4Þ

ðT  20Þ2:6  108

NACFST RAC50FST RAC100FST

0.8

RSIrc

578

1.0 0.9

g ¼ 0:84244  0:08801r þ 0:11ðT  20Þ  102 200 C 6 T 6 700 C

0.8

where RSIr refers to the modified estimation of the RSI for RACFST stub columns and r is the RCA replacement ratio. g is proposed for modifying the existing RSI formula.

0.7

Ke

ð5Þ

0.6 0.5

3.4. Elastic modulus In this study, the elastic modulus (Es) of the RACFST stub column was defined as the secant modulus while stress was equal to 40% of the peak stress [34]. Like compressive strength, the residual elastic modulus ratio Ke was also proposed to reflect the decrease of the elastic modulus after elevated temperature exposure [31] and was defined as Eq. (6).

Ke ¼

Es ðTÞ Es ð20Þ

0.3 0.2

200

500

700 o

High temperature ( C) Fig. 10. Residual elastic modulus ratio of RACFST stub columns after exposure to high temperature.

ð6Þ

where Es(T) refers to the elastic modulus of specimens after exposure to a temperature (T) and Es(20) refers to the elastic modulus of specimens at room temperature of 20 °C. As shown in Fig. 10, the Ke for the RACFST stub columns was in the ranges of 0.782–0.898 at 200 °C, 0.635–0.685 at 500 °C and

1.0 T=200 ºC

NACFST RAC50FST RAC100FST

0.8

RSIc

NACFST RAC50FST RAC100FST

0.4

T=500 ºC

0.6

0.443–0.503 at 700 °C, respectively. From Figs. 4 and 5, it is evident that there are no significant variations for Ke of RACFST stub columns with different RCA replacement ratios. It was also obvious that the Ke of the RACFST stub columns declines with the increase in high temperature, decreasing almost linearly. This effect was caused by the elastic modulus deterioration of the steel and RAC after exposure to elevated temperatures [19]. The recession mechanism was similar to the effect in strength deterioration. After exposure to certain high temperature, the material properties of steel and RAC were reduced with a consequent decrease in the elastic modulus of RACFST stub columns. Like the residual elastic modulus ratio Ke, the RCA elastic modulus ratio RCAKe was defined as shown in Eq. (7).

RCAK e ¼ 0.4

Es ðrÞ Es ð0Þ

ð7Þ

T=700 ºC 2

R =-0.477

0.2

0.0 0.0

0.2

0.4

0.6

0.8

1.0

RSIe Fig. 8. Comparison between formula calculated and experimental residual strength index of RACFST stub columns.

where Es(r) refers to the elastic modulus of RACFST stub columns with different RCA replacement ratios, and Es(0) is the elastic modulus of the corresponding NACFST column. The RCAKe of the RACFST stub columns after exposure to elevated temperature is shown in Fig. 11, where it is found that the range of RCAKe (from 0.443 to 0.912) was less than that of Ke (from 0.799 to 1.001). For instance, when temperature arrived at 700 °C, the minimum Ke value of was 0.443, while all the RCAKe values were larger than 0.799. Although there were some fluctuations,

579

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

3.5. Peak strain

1.1

To investigate the change of peak strain of RACFST stub columns after exposure to high temperature, the residual peak strain ratio (Ke) was defined as in Eq. (9).

1.0

RCAKe

0.9

Ke ¼

0.8 NACFST RAC50FST RAC100FST

0.7

0.6

200

500

700 o

High temperature ( C) Fig. 11. RCA residual elastic modulus ratio of RACFST stub columns after exposure to high temperatures.

almost all the RACFST stub columns had RCAKe less than 1.0 (the RCAKe of the NACFST column was 1.0), and it declined with the increase in elevated temperature. Furthermore, the RAC100FST stub column generally had lower RCAKe than that of the RAC50FST column. These results indicated that the RACFST stub columns exposed to elevated temperature exhibited lower elastic moduli than that of corresponding NACFST column, and this trend became more noticeable with the increase in temperature. This finding implies that, after exposure to elevated temperature, the RAC exhibited more severe decrease in elastic modulus compared to the NAC, in particular for high RCA replacement ratio. A relevant formula for calculating Ke of NACFST stub column was proposed by Han [31] as shown in Eq. (8).

K e ¼ 1  9ðT  20Þ  104

ð8Þ

where T is a high temperature to which the RACFST stub columns are exposed. The calculated Ke (Ke,c) of the RACFST stub columns is shown in Fig. 12, demonstrating that this formula, Eq. (8), can appropriately calculate the Ke of heat-treated RACFST stub columns with different RCA replacement ratios. From that comparison, the calculated values (Ke,c) slightly underestimate the experimental Ke,e, which makes this calculation formula relatively conservative for the practical structural applications.

eðTÞ eð20Þ

where e(T) represents the peak strain of RACFST stub columns after exposure to a high temperatures, and e(20) denotes the peak strain of RACFST columns that is not exposed to high temperature. The Ke for the RACFST stub columns is revealed in Fig. 13. It is noted that the distribution of the Ke of the RACFST stub columns seems different from the distribution of the RSI and Ke. Indeed, the Ke increased with the increase in high temperature. This effect might be due to damage inflicted by high temperature on the RAC and steel tubes, which consequently exhibited a decrease in stiffness and an increase in deformation. The Ke increased almost linearly with the elevated temperature, and only slight variation occurred for different RCA replacement ratios. At temperature of 200 °C, the change in range of the peak strain was not very obvious, the Ke only ranging from 1.077 to 1.173. However, when the high temperature increased to 500 °C and 700 °C, the values of Ke lay between 2.176–2.469 and 3.154–3.370, respectively. Like the residual peak strain ratio (Ke), RCA residual peak strain ratio (RCAKe) was defined as in Eq. (10), and was used to evaluate the effect of RCA replacement ratio on the peak strain of RACFST stub columns after exposure to high temperatures.

RCAK e ¼

eðrÞ eð0Þ

ð10Þ

where e(r) is the peak strain of the RACFST stub columns with RCA replacement ratios (r), and e(0) is the peak strain of the NACFST column. As shown in Fig. 14, after exposure to elevated temperatures, almost all the RCAKe of the RACFST stub columns are greater than 1.0 (the RCAKe is 1.0 for the NACFST column), and RAC100FST stub column exhibited obviously higher RCAKe than the RAC50FST column, indicating that the peak strain of the RACFST stub columns increases with the increase in the RCA replacement ratio. Previous studies have also shown that the peak strain of RAC and RACFST increases with the increase in RCA content at room temperature [35–37]. This trend was also evident for the RCAKe of RACFST stub columns suffering from exposure to elevated temperatures. Thus, RACFST stub columns tended to have better deformation capacity

4.0

1.0 T=200 ºC

0.8

ð9Þ

3.5

NACFST RAC50FST RAC100FST

3.0 2.5

0.6

Ke,e

Ke

T=500 ºC

0.4

T=700 ºC

2.0 1.5 NACFST RAC50FST RAC100FST

1.0

2

R =0.783

0.2

0.5 0.0 0.0

0.0 0.2

0.4

0.6

0.8

1.0

Ke,c Fig. 12. Comparison between calculated and experimental residual elastic modulus ratio of RACFST stub columns.

200

500

700 o

High temperature ( C) Fig. 13. Residual peak strain ratio of RACFST stub columns after exposure to high temperature.

580

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

sonably estimate the load-carrying capacity, stiffness and deformation of RACFST columns after exposure to elevated temperatures, but more experimental data and theoretical analysis are required to further validate the prediction accuracy of the modified formulas.

1.4 NACFST RAC50FST RAC100FST

1.3

RCAKe

1.2

4. Conclusions

1.1

In this study, the mechanical behaviors of RACFST stub columns under compression with RCA replacement ratios of 0, 50%, and 100% were experimentally investigated after exposure to elevated temperatures of 200 °C, 500 °C and 700 °C. The main conclusions that can be drawn are as follows:

1.0 0.9 0.8

200

500

700 o

High temperature ( C) Fig. 14. RAC residual peak strain ratio of RACFST stub columns after high temperature exposure.

than the corresponding NACFST column. This effect was probably due to the higher cement mortar content in the RAC. Moreover, the RCAKe here was unlikely to change monotonously with the increase in elevated temperature as did that of RCARSI. It also seems that the range of RCAKe was small when compared with that of Ke. As for the estimations of compressive strength and elastic modulus of RACFST stub columns after exposure to elevated temperatures, a regression formula based on the experimental results was proposed to evaluate the variation of the Ke of RACFST stub columns after exposure to elevated temperatures. The regression formula, Ker, here is shown in Eq. (11). As shown in Fig. 15, excellent agreement between the calculated value (Ker,c) and experiment result (Ker,e) is obtained.

K er ¼ 0:41814  0:08659r þ 0:418ðT  20Þ  102 200 C 6 T 6 700 C

ð11Þ

where r is the RCA replacement ratio of the RACFST stub columns. Although comprehensive experimental investigation has been conducted on the compressive mechanical behaviors of RACFST columns after exposure to high temperatures, there is a further necessary to conduct theoretical analysis to derive relevant formulas for calculating load capacity, stiffness and deformation resistance of RACFST columns under uniaxial compression in the future. On the other hand, the current modified formulas can rea-

4.0 3.5 NACFST RAC50FST RAC100FST

3.0

T=700 ºC

K ,e

2.5

(1) After exposure to elevated temperatures, there were no obvious differences between the failure mode of the RACFST stub columns, but their color changed with the increase in elevated temperature. The failure patterns of the elevated temperature heated RACFST stub columns with different RCA replacement ratios, were generally similar, manifested local outward folding and buckling of the outer steel tube. (2) The formations of load-strain curves of elevated temperatures heated RACFST stub columns incorporating different RCA replacement ratios were generally similar to each other. However, with the increase in elevated temperature, the elastic stage became progressively shorter, while the elastic-plastic stage turned progressively longer. The deformation and ductility increased with the increase in temperature. (3) The compressive strength of the RACFST stub columns declined dramatically, when they were exposed to elevated temperatures of 500 °C and 700 °C. The mechanical strength of the RACFST stub columns after heat exposure also decreased along with the increase of the RCA replacement ratio. The regression formula of the residual strength index seemed to underestimate the residual strength of the RACFST columns. (4) The elastic modulus of the RACFST stub columns decreased linearly with the increase in elevated temperature. The elevated temperature heated RACFST stub columns with higher RCA replacement ratios usually exhibited lower elastic modulus than the NACFST column. The regression formula for calculating the residual elastic modulus ratio of the NACFST stub column after exposure to high temperatures was appropriate for the RACFST columns. (5) The RACFST stub columns showed higher peak strain than that of the corresponding NACFST column. The peak strain of the RACSFST stub columns increased with the increase in elevated temperatures. However, the effect of the RCA replacement ratio on the peak strain ratio was inconsistent under different elevated temperature exposure. Application of the proposed formula for the peak strain ratio produced good consistency with the experimental results.

T=500 ºC

2.0

Acknowledgements

1.5 2

R =0.98

1.0

T=200 ºC

0.5 0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

K r,c Fig. 15. Comparison of formula calculated experimental residual peak strain ratio of RACFST stub columns.

The authors gratefully acknowledge the financial supports of the Australian Research Council Discovery Early Career Researcher Award (DE150101751) and Australian Research Council Research Hub for Nanoscience Based Construction Materials Manufacturing (IH150100006). The authors are also grateful for the financial supports of the National Natural Science Foundation of China (51408210) and University of Technology Sydney grants (Blue Sky Research Scheme and Early Career Researcher), Australia.

W. Li et al. / Construction and Building Materials 146 (2017) 571–581

References [1] J. Xiao, W.G. Li, Y. Fan, X. Huang, An overview of study on recycled aggregate concrete in China (1996–2011), Constr. Build. Mater. 31 (2012) 364–383. [2] N. Kisku, H. Joshi, M. Ansari, S.K. Panda, S. Nayak, S.C. Dutta, A critical review and assessment for usage of recycled aggregate as sustainable construction material, Constr. Build. Mater. 131 (2017) 721–740. [3] K. Konno, Y. Sato, Y. Kakuta, M. Ohira, The property of recycled concrete column encased by steel tube subjected to axial compression, Trans. Jpn. Concr. Inst. 19 (1997) 231–238. [4] R. Abendeh, H.S. Ahmad, Y.M. Hunaiti, Experimental studies on the behavior of concrete-filled steel tubes incorporating crumb rubber, J. Constr. Steel Res. 122 (2016) 251–260. [5] E.K. Mohanraj, S. Kandasamy, R. Malathy, Behaviour of steel tubular stub and slender columns filled with concrete using recycled aggregates, J. South Afr. Inst. Civ. Eng. 53 (2) (2011) 31–38. [6] W.G. Li, J. Xiao, C. Shi, C.S. Poon, Structural behaviour of composite members with recycled aggregate concrete – an overview, Adv. Struct. Eng. 18 (6) (2015) 919–938. [7] Y.F. Yang, L.H. Han, Experimental behaviour of recycled aggregate concrete filled steel tubular columns, J. Constr. Steel Res. 62 (12) (2006) 1310–1324. [8] C. Gao, L. Huang, L.B. Yan, B. Kasal, W.G. Li, Behavior of glass and carbon FRP tube encased recycled aggregate concrete with recycled clay brick aggregate, Compos. Struct. 155 (2016) 245–254. [9] V.W. Tam, Z.B. Wang, Z. Tao, Behaviour of recycled aggregate concrete filled stainless steel stub columns, Mater. Struct. 47 (1–2) (2014) 293–310. [10] L.H. Han, Y.F. Yang, H. Yang, J.S. Huo, Residual strength of concrete-filled RHS columns after exposure to the ISO-834 standard fire, Thin-Walled Struct. 40 (12) (2002) 991–1012. [11] L.H. Han, J.S. Huo, Y.C. Wang, Compressive and flexural behaviour of concrete filled steel tubes after exposure to standard fire, J. Constr. Steel Res. 61 (7) (2005) 882–901. [12] Fire Resistance Tests-elements of Building Construction, International Standard ISO 834, Geneva, French, 1975. [13] H. Abbas, Y. Al-Salloum, S. Alsayed, M. Alhaddad, R. Iqbal, Post-heating response of concrete-filled circular steel columns, KSCE J. Civ. Eng. (2016), http://dx.doi.org/10.1007/s12205-016-0852-3. [14] S.F. Jiang, Z.Q. Wu, D.S. Niu, Experimental study on fire-exposed rectangular concrete-filled steel tubular (CFST) columns subjected to bi-axial force and bending, Adv. Struct. Eng. 13 (4) (2010) 551–560. [15] G.B. Dinh, Y.C. Wang, Finite element studies of creep behavior of concretefilled steel tubular columns under compression at high temperatures, Adv. Sci. Lett. 19 (2) (2013) 508–512. [16] Y. Yao, X.X. Hu, Cooling behavior and residual strength of post-fire concrete filled steel tubular columns, J. Constr. Steel Res. 112 (2015) 282–292. [17] DBJ13-51-2003, Technical Specification for Concrete-filled Steel Tubular Structures, Ministry of Construction, Fujian Province, China, 2003 (in Chinese). [18] SEI/ASCE/SFPE 29-99, Standard Calculation Methods for Structural Fire Protection, American Society of Civil Engineers Reston, VA Society of Fire Prevention Engineers, 2003.

581

[19] Y.F. Yang, R. Hou, Experimental behaviour of RACFST stub columns after exposed to high temperatures, Thin-walled Struct. 59 (4) (2012) 1–10. [20] GB50204-2002. Code for Acceptance of Constructional Quality of Concrete Structures, Beijing, China, 2002 (in Chinese). [21] ACI Committee 318, Building Code Requirements for Structural Concrete (ACI 318-05) and Commentary (318R-05), American Concrete Institute, Farmington Hills, MI USA, 2005. [22] W.G. Li, Z.Y. Luo, C. Long, C.Q. Wu, W.H. Duan, S.P. Shah, Effects of nanoparticle on the dynamic behaviors of recycled aggregate concrete under impact loading, Mater. Des. 112 (2016) 58–66. [23] M.B. Leite, P.J.M. Monteiro, Microstructural analysis of recycled concrete using X-ray microtomography, Cem. Concr. Res. 81 (2016) 38–48. [24] L.H. Han, X.L. Zhao, Z. Tao, Tests and mechanics model for concrete-filled SHS stub columns, columns and beam-columns, Steel Compos. Struct. 1 (1) (2001) 51–74. [25] J.S. Huo, Y.M. He, L.P. Xiao, B.S. Chen, Experimental study on dynamic behaviours of concrete after exposure to high temperatures up to 700 °C, Mater. Struct. 46 (1–2) (2012) 255–265. [26] G.G. Mohamedbhai, Effect of exposure time and rates of heating and cooling on residual strength of heated concrete, Mag. Concr. Res. 38 (136) (1986) 151– 158. [27] S.H. Cai, Modern Concrete Filled Steel Tube Structure, China Communication Press, Beijing China, 2007 (in Chinese). [28] L.H. Han, J.S. Huo, Concrete-filled hollow structural steel columns after exposure to ISO834 fire standard, J. Struct. Eng. 129 (1) (2003) 68–78. [29] B. Georgali, P.E. Tsakiridis, Microstructure of fire-damaged concrete: a case study, Cem. Concr. Compos. 27 (2) (2005) 255–259. [30] D. Cree, M. Green, A. Noumowé, Residual strength of concrete containing recycled materials after exposure to fire: a review, Constr. Build. Mater. 45 (7) (2013) 208–223. [31] L.H. Han, Concrete-Filled Steel Tubular Structures-Theory and Practice, China Science Press, Beijing, 2007 (in Chinese). [32] J.B. Vieira, J.R. Correia, J.D. Brito, Post-fire residual mechanical properties of concrete made with recycled concrete coarse aggregates, Cem. Concr. Res. 41 (5) (2011) 533–541. [33] C. Laneyrie, A.L. Beaucour, M.F. Green, R.L. Hebert, B. Ledesert, A. Noumowe, Influence of recycled coarse aggregates on normal and high performance concrete subjected to elevated temperature, Constr. Build. Mater. 111 (2016) 368–378. [34] M.X. Xiong, J.Y.R. Liew, Mechanical behaviour of ultra-high strength concrete at elevated temperatures and fire resistance of ultra-high strength concrete filled steel tubes, Mater. Des. 104 (2016) 414–427. [35] G. Wardeh, E. Ghorbel, H. Gomart, Mix design and properties of recycled aggregate concretes: applicability of Eurocode 2, Int. J. Concr. Struct. Mater. 9 (1) (2015) 1–20. [36] J.Z. Xiao, W.G. Li, Z.H. Sun, S.P. Shah, Crack propagation in recycled aggregate concrete under uniaxial compressive loading, ACI Mater. J. 109 (4) (2012) 451– 461. [37] Y. Wang, J. Chen, Y. Geng, Testing and analysis of axially loaded normalstrength recycled aggregate concrete filled steel tubular stub columns, Eng. Struct. 86 (2015) 192–212.