Mechanical strength and drying shrinkage properties of concrete containing treated coarse recycled concrete aggregates

Construction and Building Materials 68 (2014) 726–739

Contents lists available at ScienceDirect

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

Mechanical strength and drying shrinkage properties of concrete containing treated coarse recycled concrete aggregates Sallehan Ismail ⇑, Mahyuddin Ramli School of Housing, Building and Planning, Universiti Sains Malaysia, 11800 Pulau Pinang, Malaysia

h i g h l i g h t s  We studied method for improving the properties of recycled concrete aggregate (RCA).  A combination of two surface treatment methods were applied on coarse RCA.  Effect of treatment methods significant to enhance the properties of coarse RCA.  Treatment methods remarkably affect mechanical strength and drying shrinkage of concrete.

a r t i c l e

i n f o

Article history: Received 5 February 2014 Received in revised form 8 May 2014 Accepted 29 June 2014

Keywords: Recycled concrete aggregate Surface treatment method Calcium metasilicate Hydrochloric acid Mechanical strength Dry shrinkage

a b s t r a c t In recycling concrete, the crushing process leaves weak mortar particles and surface cracks throughout the recycled concrete aggregates (RCA). Thus, the process is detrimental, resulting in inferior aggregate properties. This experimental study presents a method to improve the properties of coarse RCA by modifying their surface structure through the combination of two different surface treatment methods. In this study, coarse RCA are first treated by soaking in hydrochloric (HCl) acid at 0.5 mol (M) concentration. They are then impregnated with calcium metasilicate (CM) solution to coat their surface with CM particles. The effects of both surface treatments on the properties of RCA before and after treatment are determined. Moreover, the effect of the replacement of natural coarse aggregates with 60% treated coarse RCA on the mechanical strength of concrete is evaluated. The findings of this study show that the effect of the combination of these two surface treatment methods is beneficial, as the combined methods not only modify RCA surface but also enhance RCA properties. More specifically, after treatment, the particle density, water absorption, and mechanical strength of RCA are significantly improved. Consequently, the incorporation of treated RCA in concrete results in a mechanical strength that approximates concrete prepared with natural aggregates and surpasses the strength of concrete prepared with untreated RCA. In addition, the effect use of treated RCA tends to reduce the drying shrinkage of concrete. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction The recycling of concrete waste into recycled concrete aggregates (RCA) has been identified as a potential source of construction aggregates. Previous studies have highlighted the benefits of large-scale recycling of concrete waste: it reduces the quantity of concrete waste that otherwise would have been disposed in landfills, decreases the dependence of the construction industry on natural aggregates, thereby preserving natural resources, provides savings from the treatment of waste disposal, and yields alternative sources for urban areas facing shortage of natural aggregates [1–5]. Moreover, given the urgent need to ⇑ Corresponding author. Tel.: +60 0123707810; fax: +60 046576523. E-mail address: [email protected] (S. Ismail). http://dx.doi.org/10.1016/j.conbuildmat.2014.06.058 0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.

preserve the environment and maintain ecological balance, the method ensures sustainable development. The most common application of RCA is in the manufacture of roadbed gravel rather than concrete [6]. However, the acceleration of urbanization as a result of population growth has led to a greater demand for concrete. Concrete is one of the most important construction materials, with an estimated annual worldwide production of about five billion tonnes [7]. This situation requires considerable quantities of natural aggregate resources, the single largest component of concrete, making up 70–80% of its total volume [8]. However, the construction industry still has misgivings on the use of RCA in the commercial production of concrete, especially in structural applications. This factor is attributable to certain unfavorable qualities of RCA compared with those of natural aggregates. RCA is produced by

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

crushing concrete lumps into smaller particles, which are then separated using a sieve of specific size. This conventional crushing technique, such as the use of a jaw crusher, leaves particles of old mortar (cement paste) in the original aggregate particles of RCA. The amount of old mortar incorporated in RCA varies across different reports, but it can reach as high as 56% [9]. The presence of old mortar particles, which are characterized by relatively high porosity [10–14] results in the inferior quality of RCA compared with natural aggregates [15,16]. Moreover, the impact stress caused by the crushing process makes the surface layer of RCA weak, porous, and brittle [17]. The process also leaves numerous microcracks in RCA [12]. Thus, compared with natural aggregates, RCA are characterized by lower density, lower specific gravity, higher water absorption, and higher porosity [10,11,16,18,19]. These properties of RCA reportedly account for the decrease in the compressive, flexural, and tensile strength, as well as in the elastic modulus of concrete prepared with RCA [20–26]. Most researchers agree that the presence of weaker and porous old mortar in RCA particles is the main reason for the adverse characteristics of RCA and the overall deterioration of the mechanical strength of concrete. In terms of the microstructure of concrete, the interface zone between the aggregate and the cement paste is important because this zone governs the mechanical strength of concrete [27]. Using scanning electron microscopy (SEM), Katz [28] found that the surface of RCA crushed by a jaw crusher are covered with loose particles, which may lower the bond between the RCA and new cement mortar, leading to a decrease in the mechanical strength of concrete. Tam et al. [12] similarly concluded that the adherence of old mortar composed of many minute pores and cracks on RCA results in the weakening of the links in the microstructure of concrete and ultimately affects the strength of concrete. Moreover, the high porosity and water absorption of RCA leads to a decrease in the effective water content for the hydration process and consequently results in a loose interfacial transition zone (ITZ) between the RCA and the new mortar in the hardened concrete [29]. Despite these disadvantages, the use of RCA in the production of concrete is still of particular interest because of the other economic and environmental benefits it offers. As such, various approaches and methods of treatment have been developed and studied to improve the material and minimize its disadvantages. Surface treatment is an innovative and beneficial method, which modifies and enhances the physical properties of RCA before its use in the concrete mix. The literature indicates various procedures of surface treatment in RCA. For instance, Tam et al. [30] proposed the use of a low concentration of acid to minimize weak or loose mortars attached on the surface of RCA particles, thereby improving the surface contact between the aggregate and the cement mortar. In this method, RCA is soaked in three different types of acid, namely, hydrochloric acid (HCl), sulfuric acid, and phosphoric acid, at a molarity of 0.1 M for 24 h. In general, the treatment significantly reduces the water absorption of RCA by 7.27–12.17%. As a result, the compressive strength, flexural strength, and elastic modulus of the treated RCA are improved compared with those of untreated RCA. Another procedure of RCA surface treatment is the modification or improvement of the surface of RCA by refilling the pores and cracks using suitable mineral admixtures like microfillers. Katz [28] introduced the surface treatment technique by impregnating RCA with a silica fume (SF) solution. In this method, the dried RCA is soaked in the silica fume solution to coat the surface of the RCA with the silica fume particles. This treatment strengthens the structure of the aggregate, particularly the ITZ between the RCA surface and the cement paste, thus improving the mechanical strength of the concrete. Other alternative methods reported include treatment by soaking in other types of admixtures or

727

solutions, such as nanosilica solution [31], polymer solution [11] and silane-based water repellent [32]. Although each method implies a different and novel approach, they all improve the physical properties of RCA and minimize its adverse effects on concrete. However, the effects of combining the methods on the performance of the resulting concrete are not yet fully known. As such, this study aims to evaluate the feasibility of improving the poor physical properties of coarse RCA using a dual treatment method, in which coarse RCA are first pre-soaked in an acid solvent and then impregnated with a mineral admixture solution. The mineral admixture used is calcium metasilicate (CM), which eventually forms the coating layer of the impregnated RCA. This study investigates the effectiveness of this combined and dual treatment in enhancing the properties of coarse RCA and the mechanical strength of the resulting concrete. 2. Materials 2.1. Cement Type I ordinary Portland cement (Cement Industries of Malaysia Berhad) with a specific gravity of 3.15 g/cm3 was used as the main binder for the experiment. Its typical chemical compositions are presented in Table 1. 2.2. Aggregates In this study, all the coarse aggregates used had a maximum size of 20 mm. The natural coarse aggregates used were crushed granite. The coarse RCA used were generated from waste concrete cubes collected from the debris area of the Laboratory School of Housing, Building, and Planning, USM Penang, Malaysia. The strength of the waste concrete cube was unknown. The concrete cubes were first crushed and then further pounded using a steel hammer to reduce their sizes. The concrete lump was placed in a jaw crusher, where it was broken down into smaller particles. After the crushing process, the RCA were graded according to particular sizes using a vibrator sieve to obtain the size required. The fine aggregates used were uncrushed quartzite natural river sand. The aggregates were washed with water to remove any unwanted substances such as clay, dirt and dust, after which they were air-dried. The gradation of the coarse and fine aggregates based on the sieve analysis is presented in Table 2. 2.3. Superplasticizer and mixing water To enhance the workability of the concrete, a chloride-free super plasticizing admixture based on sulfonated naphthalene polymers was used. This superplasticizer complied with BS 5075-3 [33]. The mixing water used was tap water. 2.4. Acid In the present study, the method involves the application of hydrochloric acid (HCl) as an acidic solvent in the degradation action for the removal of crumbs or loosely adhered mortars attached to the original RCA aggregate. Selection with HCl resulted in improved properties of the recycled aggregate concrete and marked improvement after pre-treatment, as reported by Tam et al. [30], which is due to its effectiveness in the treatment of RCA. In addition RCA subjected in acid solution was used to remove the adhered mortar and did not corrode or interfere with the original aggregate of RCA. Given that granite is a common natural coarse aggregate used for the production of concrete in Malaysia, the RCA produced in this study is predominantly composed of granite as the natural aggregate. Hence, selecting HCl is considered suitable for removing RCA mortar because of the highly corrosive-resistant nature of granite even at high acid concentrations [34]. Nevertheless, the prepared acid solution used for treating RCA has a low concentration with a molarity of 0.5 M. This optimal concentration is in accordance with the findings recommended by the works of Ismail and Ramli [35]. The HCl used in this research is supplied by the School of Chemistry, USM Penang, Malaysia. 2.5. Calcium metasilicate (CM) CM or wollastonite, whose molecular formula is CaSiO3, is widely used in the production of ceramics, insulation, roof tiles, and other construction materials. The CM used in this study was supplied by Berjaya Bintang Timur Sdn Bhd. It came in the form of white powder, with the range of its particle sizes similar to that of the cement particles. The specific gravity of CM is 2.87 g/cm3 and its loss on ignition is 0.46%, values that are lower than those of ordinary Portland cement. The chemical composition of CM is presented in Table 1. The major compounds present in CM include 50.3% SiO2 and 44.4% CaO. Fig. 1 shows the X-ray diffraction (XRD) pattern

728

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

Table 1 Chemical composition of cement and calcium metasilicate. Material

Chemical composition (%)

Cement Calcium metasilicate (CM)

SiO2

Al2O3

Fe2O3

CaO

MgO

K2O

P2O5

MnO

TiO2

Others

16 50.32

3.6 0.77

2.9 0.33

72 44.44

1.5 1.31

0.34 0.15

0.06 0.08

0.03 0.05

0.17 0.03

3.41 2.52

LOI

Specific gravity (g/cm3)

2.53 0.46

3.15 2.87

Table 2 Sieve analysis of aggregates. Fine modulus

0.3

0.6

1.18

2.36

5

10

14

20

0.9 0.0 0.0

8.8 0.0 0.0

22.7 0.0 0.0

45.3 0.0 0.0

77.4 0.2 0.4

100 0.2 0.8

100 31 30

100 59.2 60.4

100 100 100

Wollastonite

Quartz

Sand Coarse natural Coarse RCA

Aggregate passing (%) according to sieve size (mm) 0.15

Calcite

Aggregate

Fig. 1. XRD pattern of calcium metasilicate.

of CM. Based on the XRD pattern, the major mineral phases of the CM include wollastonite (CaSiO3), quartz (SiO2), and calcite (CaCO3). Fig. 2 shows the SEM of CM, which showed an acicular form or needle-like structure. Given its acicular nature, CM can potentially be used as microfibers to reinforce cementitious materials [36–40]. Mathur, Misra, and Goel investigated the application of CM in concrete mixes as a partial replacement for cement and sand, and found that CM significantly reduced abrasion loss and shrinkage and enhanced the durability of concrete [41]. In another study, the inclusion of CM as partial replacement for cement reportedly improved the early-age engineering performance of ultra high-performance concrete [42]. Kalla et al. [43] found that the incorporation of wollastonite–fly ash in concrete mixes possibly enhanced the mechanical property and durability of concrete. However, available data for experiments using CM as materials for the surface treatment of RCA have not yet been examined. In addition, information on the use of CM in the production of recycled aggregate concrete is lacking.

3. Experimental program 3.1. Surface treatment of coarse RCA This experiment involves the combination of two different methods to improve the quality of the coarse RCA.

Fig. 2. SEM of calcium metasilicate.

3.45 6.17 6.08

729

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

Aggregate

Loose cement paste

(a)

Aggregate

Cement paste

used to coat the RCA surface would be dissolved during mixing and are expected to function as a filler with the product of cement hydration for the densification of the interface structure, which improves bond strength at contact between the aggregate surface and the cement matrix. The procedures involved in this stage included the following: (i) The coarse RCA was completely dried in an oven for 24 h at 105 °C and then cooled at room temperature. (ii) The CM solution was prepared by dissolving 10% weight CM (in powder form) in distilled water. The mixture was stirred for several minutes to ensure proper dispersion of the CM particles. (iii) The dried coarse RCA was then immediately added to the solution. The quantity of coarse RCA added was based on a weighted waterto-coarse RCA ratio of 1:1.7. (iv) The RCA particles were soaked in the CM solution for 24 h to increase CM absorption. (v) After immersion, the aggregates were drained for 10 min. A certain amount of CM particles composed of sediments at the bottom of the container was mixed by hand to evenly coat the surface of the aggregate particle. (vi) The RCA was dried in an oven at 105 °C for 24 h. The aggregate was cooled to room temperature before being used in the concrete mixture. As observed, the percentage of the CM particles that adhered on the surface of the RCA was approximately 80%. The remainder of the CM particles did not refill nor adhere on the RCA surface (i.e., in loose particle form) or were lost during the draining of the treatment solution. SEM was used to visualize the microstructures on the surface of the treated RCA. The images of the treated RCA after CM impregnation are shown in Fig. 4. The surface-treated RCA was covered with a layer of small particles identified as CM particles. 3.2. Concrete specimen preparation and curing

(b) Fig. 3. SEM images of the surfaces of (a) untreated RCA and (b) RCA after subjected to 0.5 M HCl.

3.1.1. First stage: acid soaking of RCA The SEM images in Fig. 3a show that the surface of untreated RCA is considerably more porous and is covered with a certain amount of loose cement paste (crumbs) and other small impurities, such as dust, which are loosely connected to the bulk aggregate of RCA particles as a result of the crushing process. In this stage of the treatment, the coarse RCA were immersed in HCl for the removal of weak or loose mortar particles adhering to the original RCA aggregates. Following Tam et al. [30] the treatment involved the following procedures: (i) The coarse RCA were placed in a plastic container. (ii) HCl with a molarity of 0.5 M was added until the surface of the coarse RCA was covered. The aggregates were kept immersed in acidic solvents for 24 h. (iii) The container was occasionally shaken to ensure a more efficient reaction of the acid in the degradation of weak mortar. (iv) After the immersion, the aggregates were watered with distilled water and drained. The sample was passed through a 5.00 mm sieve to ensure that only the coarse aggregates were retained. Fig. 3b shows the SEM images of the surface of the RCA, which became cleaner and more uniform compared with untreated RCA after subjected to the acid. 3.1.2. Second stage: coating with CM After soaked in acid, the coarse RCA were impregnated with CM solution. The purpose of this step was to coat the surface of coarse RCA with CM particles to refill the pores and cracks throughout its physical surface. Simultaneously, the present CM particles that was

The proportion of the concrete mixture was designed using the Department of Environment (DOE) method, which is based on constant effective water/cement ratio of 0.41 for all concrete mixtures, to achieve a target slump range of 30–60 mm and a compressive strength of 50 MPa on the 28th day [44]. Three types of mixtures were prepared using different types of coarse aggregates. The specimen types CON composed the control concrete, which was prepared using only natural coarse aggregates. Other specimens, such as types NR60 and TR60, were composed of untreated and treated coarse RCA, respectively. The dosage compositions of the coarse aggregates in both types were designed by replacing the natural coarse aggregate with untreated and treated RCA at 60% weight of the total coarse aggregate content. The detailed compositions of the specimens are shown in Table 3. Considering the loss of water caused by the absorption of the aggregates during mixing,

Aggregate CM particle

Fig. 4. SEM images of surface-treated RCA after impregnation with CM.

730

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

Table 3 Details of mixing proportion. Specimen no.

Types of coarse aggregate

Cement (kg/m3)

Water (kg/m3)

CON NR60 TR60

Natural Untreated RCA Treated RCA

510 510 510

210 210 210

a superplasticizer was added in the concrete mix in type TR60 at 0.2% of the cement weight to maintain the slump of the concrete mix. The incorporated treated coarse RCA in these types of mixture are oven dried, as opposed to those in other mixes, which are air dried. All concrete mixes in this study were mixed in accordance to the sequence prescribed in BS1881-125 [45]. A drum mixer was used to prepare all the concrete mixes. For each concrete mix, 100 mm cubes were cast for the compressive strength test, cylinder samples with diameter of 150 mm and length of 300 mm were cast for static modulus of elasticity test and 100 mm  100 mm  500 mm prisms were cast for the flexural strength and nondestructive test. All hardened concrete specimens were cast in laboratory conditions, demolded at 24 h after casting, and then fully immersed in water at 25 ± 2 °C until 7, 28, 90, and 180 days of testing were reached. 4. Testing 4.1. Determination of the properties of coarse aggregates 4.1.1. Particle density, water absorption, mechanical strength, and chemical contents of aggregate Several tests were conducted to assess the effect of the surface treatment on the properties of coarse RCA. The aggregates used were also ensured as complying with the standard requirements for concrete based on BS 882 [46]. The density and water absorption of the aggregates were tested according to BS 812-Part 2 [47]. In this test, the coarse RCA was separated into two groups based on particle size: 20 and 10 mm. Aggregates of different particle sizes yielded different results in terms of density and water absorption [48]. The mechanical strength of the aggregates was assessed by conducting crushing and impact value tests following the procedures given in BS 812-Part 110 [49] and Part 112 [50], respectively. The chloride and sulfate contents of the aggregates were chemically tested on ground specimen according to the procedures stipulated in BS 812-Part 117 [51] and Part 118 [52], respectively. In addition the pH values of the aggregates were determined. The two different surface treatment processes involved in this study may have various impacts on the properties of coarse RCA after treatment. However, the corrosive characteristics of the acid solution are reported to prominently create a chemical reaction that particularly degrades the old cement mortar that remained on RCA [30,34]. As a result, the degradation effect of the acid may indirectly cause a permanent change in the entire bulk properties of the RCA particles. Hence, this reason justifies the focus on the physical, mechanical, and chemical comparison test on the properties of coarse RCA after being subjected to acid treatment. Additionally, understanding the effectiveness of acid for the removal of loose mortar particles on the RCA surface and the safety measures on the degradation and vulnerability effect have to be considered in terms of their impact on the properties of coarse RCA, which may consequently affect the aggregate’s durability and performance when incorporated in concrete. Therefore, appropriate tests should be conducted, including aggregate density, water absorption, mechanical strength, and chemical tests, after

Coarse aggregate (kg/m3) Gravel

RCA

956 382 382

– 574 574

Sand (kg/m3)

SP (%)

722 722 722

0 0 0.2

the RCA is treated with acid. The various effects of acid on RCA properties were considered. By contrast, only density and water absorption tests were conducted to determine the porosity of coarse RCA after the second surface treatment, which involved coating coarse RCA with CM. In this case, fewer tests were conducted to examine the properties of RCA after the second surface treatment in comparison with those of RCA treated with acid because the temporary presence of CM particles on the surface of coarse RCA may impair the accuracy of the testing results. Additionally and as previously mentioned, the mainly purpose of surface-coating RCA with CM particles is to provide reinforcement in cementitious materials which are expected to be beneficial in improving the surface contact in the interface between the aggregate and the matrix rather than more focusing in modify on the properties of RCA itself.

4.1.2. Determination of mortar content At present, no standard procedure for determining the amount of adhered mortar on RCA exists. As such, the method adopted in this study to determine the quantity of adhered mortar was that of De Juan et al. [16]. The experimental study of De Juan et al. [16] reported that the thermal method is significantly suitable in determining the amount of adhered mortar. Similar findings were obtained and reported by Butler et al. [53] who found that the thermal method was more effective in removing the adhered mortar, as opposed to other methods using acid and the freeze thaw method, which are also used in quantifying the amount of adhered mortar content on RCA. The brief procedure for the determination of the adhered mortar content in this study involved the following: (i) An initial weight of oven-dried (OD) sample aggregate was recorded. (ii) The aggregate sample was immersed in and saturated with water for 24 h. (iii) The sample was heated in a muffle furnace at a temperature of 500 °C for 2 h and then dipped in cold water, resulting in a thermal crash. The remaining adhered mortar was removed by hitting the sample with a rubber mallet. (iv) Finally, the sample was sifted through a 5.00 mm sieve to ensure that only coarse aggregates were retained. Once the cleaning was completed, the sample was weighed again. The weight difference between the initial and final weight of the sample represented the total amount of the adhered mortar. The amount of the mortar was expressed as the percentage mass of RCA.

4.1.3. Determination of mortar loss This test was conducted to determine the amount of the mortar loss of RCA after immersion in 0.5 M acid at 24 h. The effect of the mortar loss of coarse RCA was measured through two separate sizes of 20 and 10 mm. The procedure used to determine the loss of adhered mortar followed that of Ismail and Ramli [35].

4.1.4. Chemical compositions of coarse aggregates The chemical composition of RCA after soaked in HCl acid was determined by the X-ray fluorescence spectrometry (XRFS) model AxiosMax (PANalytical). For these tests, the aggregate samples were ground.

731

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

4.2. Determination of concrete properties 4.2.1. Slump Slump tests of fresh concrete were conducted immediately after mixing to determine the influence of the addition of treated RCA on the workability of concrete. The slump test procedure was conducted in accordance with BS EN 12390-2 [54]. 4.2.2. Mechanical strength The compressive and flexural strength tests of the concrete were in compliance with BS EN 12390-3 [55] and BS EN 12390-5 [56], respectively. The results of the tests were taken from the average of the three specimen types (concrete cube and prism) at 7, 28, 90, and 180 days. The static modulus of elasticity (E) was determined using cylinder samples with diameter of 150 mm and length of 300 mm, following BS 1881 [22]. This test was conducted on the concrete specimens at the age of 28 days.

Cylindrical steel Data acquisition system

Computer LVDT Specimen

Load Cell

4.2.3. Impact resistance test This test was conducted to gain a better understanding of the load resistance and/or total energy absorption behavior of concrete when subjected to impact loads [57–59]. This test method was based on the work carried out by Kwan et al. [60]. The impact strength specimens prepared for the test were in the form of rectangular plates with the following dimensions: 50 mm  100 mm  500 mm. The impact resistance of concrete was determined by using a fabricated, low drop-weight impact instrument. The assembly equipment consisted of two components. (1) A single steel column that functioned as the loading frame. The column had a bracket with a cylindrical steel channel that provided a slot for the drop weight (steel ball) and acted as a guide for the steel ball when it rolled. A steel rod with a small diameter was pinned across the cylindrical steel channel to hold the steel ball and to serve as a release button for the drop ball. (2) Two support beam frames with a span of 300 mm were fixed at the bottom of the impact test rig and served as holders for the test specimen. The instrumentation of the drop-weight impact tests consisted of the dynamic load cell with a capacity of 200 kg, which was attached on both support beams to measure the impact load. The test specimen was placed at the center point of the load cell. A linear variable displacement transducer (LVDT) with a range of ±50 mm was used to gauge the deflection at the mid-span of the specimens. The LVDT was mounted on a small steel plate. It was aligned at the center of the edge of the specimens to avoid collision with the steel ball during the impact test. The input energy was provided by a steel ball weighing 450 g. This ball was released and dropped from the cylindrical steel channel at a height of 800 mm. The position of the drop-weight impact was aimed at the central span of the specimen plates. Drop-weight impact tests were repeated until failure occurred. Meanwhile, data from the load cell and the LVDT were simultaneously recorded digitally in a computer by using a National Instrument data-acquisition system. The input data of the impact energy was digitally recorded, blow by blow, through the data-acquisition system, which was controlled by LabVIEW software. In the present work, data were exported to Microsoft Excel for post-analysis. Fig. 5 presents the setup of the drop-weight impact instrument with the concrete plate specimen. 4.2.4. Non destructive test Two nondestructive tests were conducted on the hardened concrete to obtain its bulk density and ultrasonic pulse velocity (UPV). The bulk density of the hardened concrete was determined at 7, 28, 90, and 180 days using the water displacement method, a confirmed and tested method prescribed by BS EN 12390-7 [61].

Support

Fig. 5. The impact load test setup.

The UPV was measured using a portable ultrasonic nondestructive digital indicating tester, in compliance with BS EN 12504-4 [62]. The UPV test was performed using a direct transmission method for each concrete specimen at different curing ages of 7, 28, 90, and 180 days. The transit time of the ultrasonic wave through the specimen was recorded. 4.2.5. Drying shrinkage test Concrete prisms measuring 100 mm  100 mm  500 mm were prepared to conduct shrinkage test. After 24 h of removal from their mold strain gauge, studs were glued by using a special adhesive and fixed on the four longitudinal surface sides of the concrete prism. Shrinkage measurements were taken by using a demountable mechanical strain gauge (DEMEC), which can read up to 0.001 mm. The change in length of the specimens was recorded when their age of curing reached 3, 7, 14, 21, 28, 56, 91, 120, 150, and 180 days. 5. Results and discussion 5.1. Properties of the coarse aggregates The effect of surface treatment on the properties of coarse RCA were analyzed and compared. Table 4 shows the test results of the properties of coarse RCA after treated with acid, together with those of natural granite and untreated RCA. Compared with untreated RCA with natural aggregates, the coarse RCA have lower quality and are more porous. The results revealed that the coarse RCA with particle sizes of 20 and 10 mm, respectively, have densities lower by 10% and 14% and water absorptions higher by 7 and 8 times compared with those of natural coarse aggregates. In terms of mechanical strength, the RCA are weaker than the natural aggregates, whereas their aggregate impact and crushing value results are higher than those of natural aggregates. These findings may be attributed to presence

732

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

Table 4 Properties of coarse aggregate. Properties of aggregate

Sizes of aggregate (mm)

% Mortar content

20 10 20 10 20 10 20 10 14 14

Natural granite

Untreated RCA

Treated RCA After acid soaking

% Mortar loss Particle density – oven dry (Mg/m3) Water absorption (%) Agg. crushing value (%) Agg. impact value (%) pH aggregate Water-soluble chlorides (%) Acid-soluble sulfates (%) Chemical compositions SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K20 P2O5 Cl SO3

After impregnation with CM

22 45

2.60 2.58 0.60 0.70 24.32 13.98 12.6 <0.01 <0.01

2.33 2.23 4.44 5.58 29.15 21.78 12.56 <0.01 0.38

2.6 2.9 2.39 2.32 3.58 4.65 28.34 19.26 12.99 0.03 0.38

72.08 0.34 14.04 2.67 0.04 0.55 1.45 3.38 4.63 0.13 0.02 0.02

71.06 0.16 8.49 1.50 0.04 0.49 9.52 1.67 3.29 0.06 0.02 0.60

69.13 0.18 8.73 1.65 0.05 0.53 9.17 1.72 3.30 0.06 0.67 0.57

of weak adhered mortar components and mortar–aggregate bonds in the RCA. In determining the quantity of the mortar content of coarse RCA, the method was found to achieve an almost 100% removal of the adhered mortar from RCA particles based on visual inspection after completing the thermal heat process. Based on the thermal heat process results, approximately 22% of the measured mortar present in RCA covers the RCA particles at a fraction size of 10–20 mm. The percentage doubles at a fraction size of 5– 10 mm. These findings showed that the percentage of mortar in RCA tends to be higher at smaller fraction sizes. Similar findings have been reported previously [10,16]. However, as shown in Table 4, the properties of RCA immersed in acid slightly improved compared with those of untreated RCA. As observed from this study, an effect of low-concentration acid on RCA is the removal of a significant portion of the weak cement mortar on its surface. Approximately 3% mass of mortar was lost after RCA was immersed in acid. As a result, the particle density and mechanical strength of RCA increased compared with those of untreated RCA. Moreover, the absorption of RCA slightly decreased. An understanding of the chemical characteristics of RCA after treatment with acid is also important. An acidic environment does not lower the alkaline level of the aggregates, as indicated by the pH of the RCA treated with acid, which remained above 12. The exposure of RCA to HCl acid may put to risk the lead RCA containment. Excessive chloride leads to the corrosion of the reinforcement, thus affecting the durability of the structure. Table 4 shows the results of the chloride ion content as a percentage by mass of the aggregate to the nearest two decimal places. The test on the water-soluble chloride content of treated RCA yielded 0.03%, indicating the presence of chloride ions on this aggregate. However, the results also showed that the chloride content in treated aggregates still remained within a specified limit of 0.05%, as indicated in BS 882 [46] for valid use in normal reinforced concrete purposes. The tests on acid-soluble sulfate contents (SO3) of untreated and treated RCA yielded the same mean of 0.38% by mass on both types of RCA. The values were still below 1.0%, in

2.39 2.34 3.48 4.48

accordance with the limit set by BS 8500-2 [63]. The sulfate content of RCA was higher than that of natural aggregates as a result of the presence of hydrate cement paste; nevertheless, it is not expected to exert any detrimental influence on the cement hydration of the new concrete [13,64]. Furthermore, an XRFS analysis was conducted to identify the chemical composition of the coarse aggregates. The major oxide compounds traced in the coarse aggregates are shown in Table 4. The cause of the presence of cement paste on RCA is the percentage of CaO, which was slightly higher than that in natural aggregates. The chemical compositions of treated and untreated RCA were slightly similar. However, the slight reduction of SiO2 and CaO elements in treated RCA relative to untreated RCA may be a result of the loss of a certain amount of cement mortar after treatment with acid. The XRFS analysis showed that the chloride content of treated RCA was relatively higher compared with that of untreated RCA. However, the chloride content was still below 1%. This result contradicts that of the water-soluble chloride test. This discrepancy may be attributed to the immersion of RCA in HCl acid, in which the chloride ions absorbed by the mortar of RCA particle were not fully extracted by the water. Given this high chloride content, this treated RCA is not recommended for application in steelreinforced concrete. The total sulfate contents of both types of RCA are higher than those of natural coarse aggregates, which is similar to a line found in the results prior to the acid-soluble sulfate tests. However, the sulfate content was still below the limit of 1.0%. Again, the present cement paste on RCA is a result of the high sulfate content of RCA relative to that of natural aggregates. A low concentration of acid dissolves only a small quantity of mortar. As such, mortar is not totally removed from the original RCA. Cracks and pores still remain on the bulk of mortar left on the RCA, thereby increasing the aggregate’s absorptivity and porosity. Continuous surface treatments after the initial treatment and impregnation of coarse RCA with CM solution significantly improved RCA properties. As shown in Table 4, the particle density of RCA slightly increased, whereas its absorptivity significantly decreased after impregnation with CM solution. These changes

733

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

The workability of all concrete mixes, which determines the mobility and placeability of the concrete mix using the slump test results, is depicted in Fig. 6. The target slumps for the concrete mix are arranged from 30 mm to 60 mm. The overall slump results indicate that the workability of CON concrete is marginally better than that of concrete with RCA with a slump value of 60 mm. The replacement of coarse natural aggregates with 60% RCA tends to decrease the NR60 concrete slump by 17%. This result may be attributed to the high absorptivity of coarse RCA caused by the porous mortar attached to it, which absorbs more water during concrete mixing, thus lowering the workability of concrete [58,65]. As observed, the slump loss tends to be greater in the TR60 concrete due to the incorporation of 60% treated RCA in OD states, which absorb more water during mixing. In addition, the CM particles coating the surface of the coarse RCA also absorbed portions of free water during mixing. However, the addition of a 0.2% mass of superplasticizer to the cement can compensate for the slump loss in the TR60 concrete mixes.

Bulk Density ( kg/m2)

5.2. Slump of concrete

2410 2400 2390 2380 2370

CON

2360

NR60

2350

TR60

2340 2330 0

20

40

60

80 100 120 140 160 180 200

Curing Age (Days) Fig. 7. Bulk densities of hardened concrete mixes.

Compressive Strength (MPa)

may be a result of the coating of CM particles on the surface of RCA, which forms a protective layer and refills the pores and cracks, thereby improving the quality of the treated RCA.

5.3. Bulk density of concrete

70 60 50 40 CON

30

NR60

20

TR60

10 0 0

40

80

120

160

200

Curing Age (Days) The bulk density of hardened concrete at 7, 28, 90, and 180 days is illustrated in Fig. 7. In general, the bulk densities of the hardened concrete mixes slightly increased along with the number of curing days. As Fig. 7 indicates, among all the concrete specimens, the CON concrete has the greatest density. The lower density of concrete mixes containing coarse RCA is due to the lower particle density of coarse RCA compared with that of the natural coarse aggregates. However, the densities of the TR60 concrete mixes containing 60% treated coarse aggregates are higher than those of the NR60 concrete at ages 28, 90, and 180 days. This finding may be attributed to the reaction of CM particles with new cement paste, which increases the hydration of the cement product and makes the concrete with treated RCA denser than that with untreated RCA. 5.4. Compressive and flexural strength As Fig. 8 illustrates, all the concrete specimens show a similar profile of strength development, in which the compressive strength increases with the curing age. However, the inclusion of untreated RCA noticeably affects the compressive strength of the concrete. The results indicate that the NR60 concrete prepared with untreated RCA had a lower compressive strength than the control concrete across all testing days. Moreover, the NR60

Slump (mm)

65 60 55 50 45

CON

NR60

Specimens Fig. 6. Slump test results of concrete mixes.

TR60

Fig. 8. Development of compressive strength of concrete mixes with curing age.

Table 5 Compressive strength relative to the CON (control) concrete. Curing age (days)

CON

NR60

TR60

7 28 90 180

1 1 1 1

0.83 0.86 0.89 0.92

1.03 0.96 0.99 0.98

Table 6 Flexural strength relative to the CON concrete. Curing age (days)

CON

NR60

TR60

7 28 90 180

1 1 1 1

0.97 0.88 0.90 0.87

0.96 0.91 0.97 0.95

concrete did not achieve the target compressive strength of 50 MPa at the age of 28 days. Table 5 shows that at 7, 28, 90, and 180 days, the strength of the NR60 concrete prepared was 83%, 86%, 89%, and 92% of the control concrete, respectively. However, the concrete prepared with the treated RCA performed better than that prepared with the untreated RCA. At 7 days, the strength of the TR60 concrete was 3% higher than that of the control concrete. At 28, 90, and 180 days, the compressive strength of the TR60 concrete with treated RCA was 96%, 99%, and 98% of the control concrete, respectively, which are relatively close to the compressive strength of the control concrete. The strength of the TR60 concrete reached more than 50 MPa, which is the targeted strength at 28 days. Table 6 presents the relative flexural strength of all concrete mixtures (expressed as a percentage of that corresponding to the control concrete) and Fig. 9 plots the development of the flexural

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

Flexural Strength (MPa)

7 6 5 4 CON

3

NR60

2

TR60

1 0 0

40

80

120

160

200

Fig. 9. Development of the flexural strength of concrete mixes with curing age.

Flexural Strength (MPa)

7 6 5 y = 1.6118x - 6.6086 R² = 0.88

3 2 1 0 5.5

6

6.5

7

7.5

37.0 36.0 35.0 34.0 33.0 32.0 31.0 30.0 29.0

CON

NR60

TR60

Specimens

Water curing age (Days)

4

Stac Modulus of Elascity (GPa)

734

8

8.5

Square root of compressive strength (MPa) Fig. 10. Relationship between compressive and flexural strength.

strength across the curing age. Quite similar trends were also observed in the compressive strength of concrete whose flexural strength decreases when incorporated with untreated RCA. At 7, 28, 90, and 180 days, the percentage decrease in the flexural strength of the NR60 concrete was 3%, 12%, 10%, and 13%, respectively, compared with the control concrete. However, mixing concrete and treated RCA reduced the decrease in the flexural strength of concrete. The results in Table 6 show that the decrease in the flexural strength of the TR60 concrete was only 4%, 9%, 3%, and 5%, respectively, compared with the control concrete at 7, 28, 90, and 180 days. These findings indicate that the flexural strength of the concrete with treated RCA was better than that of the concrete with untreated RCA. In addition, the data plotted in Fig. 10 indicate a very good correlation between the compressive strength and flexural splitting strength of all the concrete mixtures, with a correlation coefficient of R-square value = 0.88. Overall, the results obtained from this investigation showed that the inclusion of 60% natural coarse aggregate with untreated RCA in concrete mixtures leads to unfavorable results in terms of compressive and flexural strength. The contributing factors to these changes include the following: (1) The adhered mortar on the RCA particle results in the lower quality of RCA relative to natural aggregates. (2) The cracks and loose residual mortar particles on the surface of the RCA particle obstruct the stronger bonds between RCA and cement paste, thus creating weak links during the addition of this kind of aggregate in the concrete phase [20,28,30]. Meanwhile, the inverse results are observable in the effects of the incorporation of treated RCA in concrete. Small but noticeable reductions in the compressive and flexural strength of the concrete are observable compared with the concrete containing untreated RCA. The enhancement of the compressive and flexural strength is attributed to the following: (1) The removal of weak and loose mortar particles on the surface of the RCA by acid significantly enhances the physical and mechanical properties of RCA. Moreover, the improvement in the quality of the surface of

Fig. 11. Static modulus of elasticity of the concrete mixes at 28 days.

RCA produces a stronger surface contact at the interfacial zone between the cement paste and the RCA, which is important in reflecting the increase in concrete strength. (2) The extent of the modification of the RCA surface that follows the impregnation of the RCA with CM solution significantly decreases the number of pores and absorptive characteristics of RCA. This result is due to the coat of CM particles that refills the pores and cracks on the surface of the old mortar of RCA. The low absorption of the RCA and the presence of the CM particle on their surface reduce the formation of accumulated water films on the surface of the aggregates, resulting in a bleeding effect. This bleeding effect creates cracks along the ITZ between the new cement paste and the aggregate [66]. (3) The fine CM particles that adhered on the outer RCA surface are dissolved during mixing and are then mainly attributed as an inert filler material that could modify and refine the pore structure, which results in a dense microstructure particularly at the interface zone between the aggregate and the cement matrix in RAC. In addition, the presence of these fine CM particles can provide nucleation sites for the precipitation of hydration products, thereby increasing the speed of the cement hydration process [67,68] and promoting the formation of a stronger and denser cement gel. Consequently, all these effects aids in strengthening the bond between the aggregate surface and cement matrix, thereby improving the mechanical strength of the RAC. 5.5. Static modulus of elasticity (E) The E of the concrete investigated at 28 days is shown in Fig. 11. This study used the same design of the cement paste for all concrete mixes. However, the effect of E was most likely influenced by the inclusion of coarse RCA. The data confirm the results obtained, namely, the E of the NR60 concrete with untreated coarse RCA decreased by almost 12% compared with the control concrete. The reduction in the E of the concrete prepared using RCA compared with that prepared using normal concrete was also observed by other researchers [58,69–71]. The reason behind this phenomenon is that the low values of the E of concrete prepared with RCA are ascribed to the porosity and low modulus of the coarse RCA compared with the natural aggregate [71,72]. According to Beshr et al. [73], the stiffness of the coarse aggregates greatly influence the E of concrete. Aïtcin and Mehta [74] also pointed out that the low strength of the aggregate may result in a concrete with a lower E. The inclusion of treated coarse RCA tends to reduce the decrease in the E of concrete. The reduction in the E of the TR60 concrete was only approximately 4% compared with that of the control concrete. The improvement observed in the E of the TR60 concrete may be attributed to the good interface bonds between the aggregate and cement paste, which helped reduce the propagation of cracks during loading. Kheder and Al-Windawi [70] and Rao

735

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

Blow 1 Blow 2

Impact Force (N)

1500

Blow 3 Blow 4

1000

Blow 5 Blow 6

500

Blow 7 Blow 8

0 -2

-1

0

-500

1

2

3

4

5

Deflecon (mm)

Energy Absorpon (kNmm)

2000

2.50 2.00 1.50 1.00 0.50 0.00

CON

NR60

TR60

28 days

1.49

1.00

1.25

90 days

1.99

1.25

1.75

180 days

2.29

2.08

2.22

(a)

Fig. 13. The energy absorbed by CON, NR60 and TR60 concrete plates under impact load.

2000

Impact Force (N)

1500 Blow 1

1000

Blow 2 Blow 3

500

CON

Blow 4 Blow 5

0 -2

-1

0

1

2

3

4

5

NR60

-500

Deformaon (mm) (b)

2000

TR60 Blow 1

Impact Force (N)

1500

Blow 2 Blow 3

1000

Fig. 14. Crack pattern of CON, NR60 and TR60 specimen.

Blow 4 Blow 5

500

Blow 6

0 -1

0

-500

1

2

3

4

5

Deflecon (mm) (c)

Fig. 12. Impact load versus midspan deflection of concrete plates at 28 days: (a) CON, (b) NR60 and (c) TR60.

et al. [18] agreed that the improvement in the bonds between the aggregate and the cement paste is an important factor that increases the E of concrete. 5.6. Impact resistance The response of the CON, NR60, and TR60 series of concrete to low-velocity impact loading was investigated at 28, 90, and 180 days. The impact resistance of the concrete plates was determined by measuring the energy absorbed by the fracture of the specimen, which included the number of blows that caused its ultimate failure. The impact energy was calculated from the area under the impact load–deflection curve [75–78] and was analyzed, moment by moment, after the steel ball was dropped. Fig. 12 shows an example of the impact load versus the mid-span,

deflection–relation curve of the specimens at 28 days. The graph shows that the curve produced a peak line, thus indicating highimpact force loading once the steel ball hit the plates. The plates continued to vibrate with small waves because some of the stored energy that was released caused the steel ball to rebound. Notably, the specimens were prepared in plain concrete. Thus, the ultimate failure on most of the concrete plates occurred simultaneously once the first visible crack formed during experimentation. The energy absorbed by the CON, NR60, and TR60 concrete plates under impact load are also shown in Fig. 13. The results show that, among all the concrete specimens, the CON concrete demonstrated the highest impact resistance of 1.49, 1.99, and 2.29 kN mm at the testing ages of 28, 90, and 180 days, respectively. However, the impact resistance of the concrete specimens with untreated, coarse recycled concrete aggregate (RCA) was lower. For example, the NR60 concrete exhibited an impact resistance of 67%, 63%, and 91% at 28, 90, and 180 days, respectively, compared with the CON concrete. The results confirmed that the stiffness of the material is significantly related to the impact resistance of concrete [58,77] because coarse aggregates are considered as dominant constituents of materials during the formation of the final concrete structure. As discussed in previous sections, the impact value of coarse RCA is less than that of coarse natural aggregates. Thus, coarse RCA contributes to reducing the mechanical strength and modulus of elasticity of concrete. In addition, the weak interface between the aggregates and the cement paste

736

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

may contribute to fracture propagation in the concrete [58]. By contrast, the resistance of concrete to crack propagation under an impact force is enhanced, as indicated in the TR60 concrete specimens. Although the energy absorption of the TR60 concrete specimens remained lower than that of the CON concrete specimens at all testing ages, the former exhibited an increment in their fracture energy compared with the NR60 concrete specimens. This phenomenon could be attributed to the improvement of the properties of coarse RCA and the modification that strengthened the interface between the aggregate–paste bond as a result of surface treatment. An examination of the surface cracks of the specimens subjected to repeated impact loads (Fig. 14) showed that the crack path patterns are more tortuous in the CON concrete. The stiffness of gravel with high stress capability prevented crack propagation. Hence, cracks failed to deviate from the interface of the particle and the matrix (see Fig. 15a). By contrast, the failure crack path pattern was even in the case of the NR60 concrete (Fig. 14), whereas the crack ruptured a few grains of coarse RCA and the

interfaces (Fig. 15b). Similar crack patterns were observed in the TR60 specimens (Figs. 14 and 15c). However, the improvement in the aggregate matrix bond and the inclusion of the treated coarse RCA in the TR60 specimens decreased the fracture energy of the concrete because of the effect of aggregates with low stiffness and quality. In this study, the effects of including untreated and treated coarse RCAs exhibited a similar behavior in influencing the static flexural load and impact load resistance (energy absorption) of concrete. Fig. 16 shows the linear relationship between these two parameters, which presents a strong correlation (R2 = 0.82). 5.7. UPV results The correlation between the compressive strength of all the concrete mixtures produced by the propagation of UPV transmitted through the concrete is shown in Fig. 17. All the concrete specimens showed very good correlation between the two variables as the R-square value of the CON, NR60, and TR60 concrete achieved

Crack failure at interface Adhered mortar

Original aggregate

Crack failure at adhered mortar of RCA

Crack failure at interface

(a)

(b)

Adhered mortar

Original aggregate

Crack failure at adhered mortar of RCA

(c) Fig. 15. Failure surfaces of: (a) CON, (b) NR60 and (c) TR60.

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

6.50 y = 1.0268x + 3.7457 R² = 0.8223

Flexural Load (MPa)

6.00 5.50 5.00 4.50 4.00 0.70

1.20

1.70

2.20

2.70

Energy Absorpon (kNmm) Fig. 16. Relationship between static flexural strength and impact energy.

Compressive Strength (MPa)

65 60

y = 0.6679e 0.9485x R² = 0.97

55 50

y = 0.115e1.2824x R² = 0.97

45 40

CON

y = 0.1828e 1.208x R² = 0.90

35

NR60 TR60

30 25 20 4.40

4.50

4.60

4.70

4.80

4.90

5.00

737

vulnerabilities to corrosion and deterioration of the reinforcement bar, particularly in structural concrete applications. Specifying the dry shrinkage of concrete in long-term performance is highly required. The Australian Standard (AS 3600 – Concrete Structures) has recommended that the nominal shrinkage limit of concrete at 56 days should not exceed 700 microstrains [79]. Fig. 18 presents the typical variations of shrinkage development with respect to time for all concrete mixes. From the test results, we observed that the dry shrinkage values of all specimen, which were measured up to 180 days overall, invariably fell below 500 microstrains. From the results during the early ages up to 28 days, the drying shrinkage behavior of all three concretes mixes were observed to approximate one another and exhibit an extremely steep development. After 28 days, however, different drying shrinkage values became clearer. In particular, NR60 concrete exhibited a high shrinkage value compared with the control concrete. When the shrinkage strain after 180 days of drying was considered, the results showed that the drying shrinkage magnitude of the NR60 concrete was 26% higher than that of the control concrete. The high shrinkage strain of the NR60 concrete was likely related to the poor quality and low stiffness of untreated, coarse RCA, which is affected in the long run. The volume and stiffness of the aggregates are considered as important factors that prevent the shrinkage of concrete [80]. However, enhancing the quality of treated RCA after surface treatment positively affected and slowed down drying shrinkage stress. 6. Conclusion

Ultrasonic Pulse Velocity (km/s) According to the results acquired throughout the experiment, the effects of surface treatment on RCA properties and on the mechanical strength of concrete can be summarized as follows:

Fig. 17. Relationship between compressive strength and UPV.

Drying Shrinkage (Micro Strain)

600 500 400 300 CON

200

NR60 TR60

100 0 0

20

40

60

80

100

120

140

160

180

200

Drying Time (Days) Fig. 18. Drying shrinkage of concrete mixtures versus drying time.

0.97, 0.90, and 0.97, respectively. Among the concrete specimens, the control concrete demonstrated higher UPV values. The UPV values of the concrete mixtures with untreated RCA were the lowest. By contrast, the UPV values gained by the TR60 concrete were slightly higher than those of the NR60 concrete. This finding can also be attributed to the improvement of RCA properties, which affect the CM particles, thus improving the microstructure of ITZ and enhancing the bond strength between the new cement paste and the RA. 5.8. Drying shrinkage The change in volume of the concrete due to shrinkage can result in the cracking of the concrete, thereby exposing

1. The combination of the two methods in the surface treatment of RCA leads to different effects on the enhancement of RCA properties after the crushing process. The first treatment used a low concentration of acid treatments, which effectively removed a certain portion of the weak cement mortar and loose substances from the RCA surface, thereby significantly improving the physical and mechanical properties of RCA. The next treatment was the impregnation of RCA with a CM solution to refill the pores on the remaining bulk mortar adhering on the RCA, thus potentially reducing the porosity of RCA. 2. The chemical test determining the total chloride and sulfate contents of the treated RCA after treatment with acid is safe to use in normal reinforced concrete, where the extract results indicate that the chloride and sulfate contents are within the limits of the respective standards. However, the XRF analysis showed that the level of the contaminant chloride ions is higher in treated RCA compared with that in untreated RCA, but the difference is not more than 1%. 3. The decrease in the compressive and flexural strength of concrete containing untreated RCA is lower than that in concrete with natural aggregates at all testing dates. Moreover, the concrete containing untreated RCA did not achieve the target compressive strength of 50 MPa at 28 days of curing. 4. The concrete prepared with the treated RCA performs better than that prepared with the untreated RCA. The inclusion of treated RCA compensates for the reduction of the decrease in the compressive and flexural strength, and E of concrete. 5. The improvement in the mechanical strength of concrete containing treated RCA compared with concrete containing untreated RCA is attributable to the surface treatment, which improves the structure of the ITZ around the RCA, resulting in an improved surface contact and bond strength between the cement matrix and the aggregate.

738

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739

6. The drying shrinkage values of concrete mixtures with treated coarse RCA were lower than those of concrete mixtures with untreated coarse RCA. 7. According to the overall results, the surface treatment method proposed in this present work is considered to be a reliable new technique that can minimize the adverse effect related to the inherent low quality of RCA products. The significant improvement in the use of treated RCA, as demonstrated in this study, enables its application in structural and non-structural concrete with less detriment to concrete performance. For safety purposes and to avoid corrosion caused by chloride, however, the treated RCA can alternatively be applied in structural concretes designed with noncorrosive reinforcement materials, such as those that use fiber-reinforced polymer (FRP), for structural reinforcement. Despite this treatment method having a multi-step process, the method does not require complicated mechanical equipment and high energy consumption. Hence, this treatment process is considered cost-effective and beneficial in providing an alternative method to encourage the application of RCA in large-scale concrete production and in ensuring a feasible method to achieve sustainability in the construction industry.

References [1] Yang D-j, Hao Y-h, Wang T-c. Experimental research on recycled aggregate concrete for highway pavement. In: Heng W, Yinhai W, Jian R, Jiancheng W, editors. Tenth international conference of Chinese transportation professionals (ICCTP). Beijing (China): ASCE; 2010. p. 309. [2] Tam VWY. Comparing the implementation of concrete recycling in the Australian and Japanese construction industries. J Clean Prod 2009;17(7):688–702. [3] Marinkovic S, Radonjanin V, Malesev M, Ignjatovic I. Comparative environmental assessment of natural and recycled aggregate concrete. Waste Manage 2010;30(11):2255–64. [4] Abbas A, Fathifazl G, Isgor OB, Razaqpur AG, Fournier B, Foo S. Environmental benefits of green concrete. EIC Climate Change Technology, 2006 IEEE; 2006. p. 1–8. [5] Kou S-c, Poon C-s, Agrela F. Comparisons of natural and recycled aggregate concretes prepared with the addition of different mineral admixtures. Cem Concr Compos 2011;33(8):788–95. [6] Tam VWY, Tam L, Le KN. Cross-cultural comparison of concrete recycling decision-making and implementation in construction industry. Waste Manage 2010;30(2):291–7. [7] Abbasi T, Abbasi SA. Decarbonization of fossil fuels as a strategy to control global warming. Renew Sustain Energy Rev 2011;15(4):1828–34. [8] Mindess S, Young JF, Darwin D. Concrete. 2nd ed. New Jersey: Prentice Hall; 2003. [9] Butler L, West JS, Tighe SL. The effect of recycled concrete aggregate properties on the bond strength between RCA concrete and steel reinforcement. Cem Concr Res 2011;41(10):1037–49. [10] Padmini AK, Ramamurthy K, Mathews MS. Influence of parent concrete on the properties of recycled aggregate concrete. Constr Build Mater 2009;23(2):829–36. [11] Kou SC, Poon CS. Properties of concrete prepared with PVA-impregnated recycled concrete aggregates. Cem Concr Compos 2010;32(8):649–54. [12] Tam VWY, Gao XF, Tam CM. Microstructural analysis of recycled aggregate concrete produced from two-stage mixing approach. Cem Concr Res 2005;35(6):1195–203. [13] Tam VWY, Le KN. Aggregate testing using 2nd-, 7th- and 10th-order interpolation polynomials. Resour Conserv Recycl 2007;52(1):39–57. [14] López-Gayarre F, Serna P, Domingo-Cabo A, Serrano-López MA, López-Colina C. Influence of recycled aggregate quality and proportioning criteria on recycled concrete properties. Waste Manage 2009;29(12):3022–8. [15] Abbas A, Fathifazl G, Fournier B, Isgor OB, Zavadil R, Razaqpur AG, et al. Quantification of the residual mortar content in recycled concrete aggregates by image analysis. Mater Charact 2009;60(7):716–28. [16] De Juan MS, Gutiérrez PA. Study on the influence of attached mortar content on the properties of recycled concrete aggregate. Constr Build Mater 2009;23(2):872–7. [17] Ogawa H, Nawa T. Improving the quality of recycled fine aggregate by selective removal of brittle defects. J Adv Concr Technol 2012;10(12):395–410. [18] Chakradhara Rao M, Bhattacharyya S, Barai S. Influence of field recycled coarse aggregate on properties of concrete. Mater Struct 2011;44(1):205–20. [19] Tam VWY, Tam CM. Assessment of durability of recycled aggregate concrete produced by two-stage mixing approach. J Mater Sci 2007;42(10):3592–602.

[20] Kwan WH, Ramli M, Kam KJ, Sulieman MZ. Influence of the amount of recycled coarse aggregate in concrete design and durability properties. Constr Build Mater 2012;26(1):565–73. [21] Ann KY, Moon HY, Kim YB, Ryou J. Durability of recycled aggregate concrete using pozzolanic materials. Waste Manage 2008;28(6):993–9. [22] Otsuki N, Miyazato S-i, Yodsudjai W. Influence of recycled aggregate on interfacial transition zone, strength, chloride penetration and carbonation of concrete. J Mater Civ Eng 2003;15(5):443–51. [23] Otsuki N, Yodsudjai W, Nishida T, Yamane H. Developed method for measuring flexural strength and modulus of elasticity of micro-regions in normal and recycled aggregate concretes. Mag Concr Res 2003;55(5):439–48. [24] Ajdukiewicz A, Kliszczewicz A. Influence of recycled aggregates on mechanical properties of HS/HPC. Cem Concr Compos 2002;24(2):269–79. [25] Poon CS, Shui ZH, Lam L, Fok H, Kou SC. Influence of moisture states of natural and recycled aggregates on the slump and compressive strength of concrete. Cem Concr Res 2004;34(1):31–6. [26] Rao A, Jha KN, Misra S. Use of aggregates from recycled construction and demolition waste in concrete. Resour Conserv Recycl 2007;50(1):71–81. [27] Alexander MG, Mindess S. Aggregates in concrete London. New York: Taylor & Francis; 2005. [28] Katz A. Treatments for the improvement of recycled aggregate. J Mater Civ Eng 2004;16(6):597–603. [29] Poon CS, Shui ZH, Lam L. Effect of microstructure of ITZ on compressive strength of concrete prepared with recycled aggregates. Constr Build Mater 2004;18(6):461–8. [30] Tam VWY, Tam CM, Le KN. Removal of cement mortar remains from recycled aggregate using pre-soaking approaches. Resour Conserv Recycl 2007;50(1):82–101. [31] Kutcharlapati S, Sarkar AK, Rajamane NP. Nanosilica improves recycled concrete aggregates. January ed. NBM Media; 2011. [32] Zhu Y-G, Kou S-C, Poon C-S, Dai J-G, Li Q-Y. Influence of silane-based water repellent on the durability properties of recycled aggregate concrete. Cem Concr Compos 2013;35(1):32–8. [33] BS 5075-3. Concrete admixtures. Specification for superplasticizing admixtures. London British Standards Institution; 1985. [34] Akbarnezhad A, Ong KCG, Zhang MH, Tam CT. Acid treatment technique for determining the mortar content of recycled concrete aggregates. J Test Eval 2013;41(3). [35] Ismail S, Ramli M. Engineering properties of treated recycled concrete aggregate (RCA) for structural applications. Constr Build Mater 2013;44:464–76. [36] Low NMP, Beaudoin JJ. Flexural strength and microstructure of cement binders reinforced with wollastonite micro-fibres. Cem Concr Res 1993;23(4): 905–16. [37] Low NMP, Beaudoin JJ. The flexural toughness and ductility of portland cement-based binders reinforced with wollastonite micro-fibres. Cem Concr Res 1994;24(2):250–8. [38] Low NMP, Beaudoin JJ. Mechanical properties and microstructure of high alumina cement-based binders reinforced with natural wollastonite microfibres. Cem Concr Res 1994;24(4):650–60. [39] Low NMP, Beaudoin JJ. Mechanical properties of high performance cement binders reinforced with wollastonite micro-fibres. Cem Concr Res 1992;22(5):981–9. [40] Low NMP, Beaudoin JJ. Stability of Portland cement-based binders reinforced with natural wollastonite micro-fibres. Cem Concr Res 1994;24(5):874–84. [41] Mathur R, Misra AK, Goel P. Influence of wollastonite on mechanical properties of concrete. J Sci Ind Res 2007;66:1029–34. [42] Soliman A, Nehdi M. Effect of natural wollastonite microfibers on early-age behavior of UHPC. J Mater Civ Eng 2012;24(7):816–24. [43] Kalla P, Misra A, Gupta RC, Csetenyi L, Gahlot V, Arora A. Mechanical and durability studies on concrete containing wollastonite–fly ash combination. Constr Build Mater 2013;40:1142–50. [44] Teychenné DC, Franklin RE, Erntroy HC, Hobbs DW, Marsh BK. Design of normal concrete mixes. 2nd ed. Watford (UK): Building Research Establishment; 1997. [45] BS 1881-125. Testing concrete. Methods for mixing and sampling fresh concrete in the laboratory. London: British Standards Institution; 1986. [46] BS 882. Specification for aggregates from natural sources for concrete. London: British Standards Institution; 1992. [47] BS 812 : Part 2. Testing aggregates. Methods of determination of density. London: British Standards Institution; 1995. [48] Hansen TC, Narud H. Strength of recycled concrete made from crushed concrete coarse aggregate. Concr Int 1983;5(01):79–83. [49] BS 812-110. Testing aggregates. Methods for determination of aggregate crushing value (ACV). London: British Standard Institution; 1990. [50] BS 812-112. Testing aggregates. Method for determination of aggregate impact value (AIV). London: British Standard Institution; 1990. [51] BS 812-117. Testing aggregates. Method for determination of water-soluble chloride salts. London: British Standards Institution; 1988. [52] BS 812-118. Testing aggregates. Methods for determination of sulphate content. London: British Standards Institution; 1988. [53] Abbas A, Fathifazl G, Burkan Isgor O, Razaqpur AG, Fournier B, Foo S. Proposed method for determining the residual mortar content of recycled concrete aggregates. J ASTM Int 2008;5(1). [54] BS EN 12350-2. Testing fresh concrete. Slump-test. London: British Standards Institution; 2009.

S. Ismail, M. Ramli / Construction and Building Materials 68 (2014) 726–739 [55] BS EN 12390-3. Testing hardened concrete. Compressive strength of test specimens. London: British Standards Institution; 2009. [56] BS EN 12390-5. Testing hardened concrete. Flexural strength of test specimens. London: British Standards Institution; 2009. [57] Banthia N, Mindess S, Bentur A, Pigeon M. Impact testing of concrete using a drop-weight impact machine. Exp Mech 1989;29(1):63–9. [58] Chakradhara Rao M, Bhattacharyya SK, Barai SV. Behaviour of recycled aggregate concrete under drop weight impact load. Constr Build Mater 2011;25(1):69–80. [59] Agar Ozbek AS, Weerheijm J, Schlangen E, Van Breugel K. Dynamic behavior of porous concretes under drop weight impact testing. Cem Concr Compos 2013;39:1–11. [60] Kwan WH, Ramli M, Cheah CB. Flexural strength and impact resistance study of fibre reinforced concrete in simulated aggressive environment. Constr Build Mater 2014;63(1):62–71. [61] BS EN 12390-7. Testing hardened concrete. Density of hardened concrete. London: British Standards Institution; 2009. [62] BS EN 12504-4. Testing concrete. Determination of ultrasonic pulse velocity. London: British Standards Institution; 2004. [63] BS 8500-2. Concrete – complementary British standard to BS EN 206-1. Specification for constituent materials and concrete. London: British Standards Institution; 2002. [64] Dhir RK, Paine KA. Performance related approach to use of recycled aggregates. Oxon: Waste & Resources Action Programme (WRAP) and University of Dundee; 2007. [65] Tangchirapat W, Buranasing R, Jaturapitakkul C, Chindaprasirt P. Influence of rice husk–bark ash on mechanical properties of concrete containing high amount of recycled aggregates. Constr Build Mater 2008;22(8):1812–9. [66] Li J, Xiao H, Zhou Y. Influence of coating recycled aggregate surface with pozzolanic powder on properties of recycled aggregate concrete. Constr Build Mater 2009;23(3):1287–91. [67] Soliman AM, Nehdi ML. Effects of shrinkage reducing admixture and wollastonite microfiber on early-age behavior of ultra-high performance concrete. Cem Concr Compos 2014;46:81–9.

739

[68] Tikkanen J. A novel application of mineral powders in normal strength concrete (Doctoral dissertations). Helsnki (Finland): Aalto University; 2013. [69] Gomes M, de Brito J. Structural concrete with incorporation of coarse recycled concrete and ceramic aggregates: durability performance. Mater Struct 2009;42(5):663–75. [70] Kheder G, Al-Windawi S. Variation in mechanical properties of natural and recycled aggregate concrete as related to the strength of their binding mortar. Mater Struct 2005;38(7):701–9. [71] Etxeberria M, Vázquez E, Marí A, Barra M. Influence of amount of recycled coarse aggregates and production process on properties of recycled aggregate concrete. Cem Concr Res 2007;37(5):735–42. [72] Tabsh SW, Abdelfatah AS. Influence of recycled concrete aggregates on strength properties of concrete. Constr Build Mater 2009;23(2):1163–7. [73] Beshr H, Almusallam AA, Maslehuddin M. Effect of coarse aggregate quality on the mechanical properties of high strength concrete. Constr Build Mater 2003;17(2):97–103. [74] Aitcin PC, Mehta PK. Effect of coarse-aggregate characteristics on mechanical properties of high-strength concrete. ACI Mater J 1990;87(2):103–7. [75] Tachibana S, Masuya H, Nakamura S. Performance based design of reinforced concrete beams under impact. Nat Hazards Earth Syst Sci 2010;10(6):1069–78. [76] Adhikary SD, Li B, Fujikake K. Dynamic behavior of reinforced concrete beams under varying rates of concentrated loading. Int J Impact Eng 2012;47:24–38. [77] Al-Tayeb MM, Abu Bakar BH, Ismail H, Akil HM. Effect of partial replacement of sand by recycled fine crumb rubber on the performance of hybrid rubberizednormal concrete under impact load: experiment and simulation. J Clean Prod 2013;59:284–9. [78] Banthia N, Gupta P, Yan C. Impact resistance of fiber reinforced wet-mix shotcrete Part 1: Beam tests. Mater Struct 1999;32(8):563–70. [79] Sagoe-Crentsil KK, Brown T, Taylor AH. Performance of concrete made with commercially produced coarse recycled concrete aggregate. Cem Concr Res 2001;31(5):707–12. [80] Adam I, Sakata K, Ayand T. Influence of coarse aggregate on the shrinkage of normal and high-strength concretes. J Fac Environ Sci Technol 2001;6(1):41–5.