mechanical and durability performance of sustainable ...

MECHANICAL AND DURABILITY PERFORMANCE OF SUSTAINABLE STRUCTURAL CONCRETES: AN EXPERIMENTAL STUDY

Ciro Faella Department of Civil Engineering, University of Salerno via Giovanni Paolo II 132, 84084, Fisciano (SA), Italy e.mail: [email protected]

Carmine Lima Department of Civil Engineering, University of Salerno via Giovanni Paolo II 132, 84084, Fisciano (SA), Italy e.mail: [email protected]

Enzo Martinelli Department of Civil Engineering, University of Salerno via Giovanni Paolo II 132, 84084, Fisciano (SA), Italy e.mail: [email protected], tel. +39 089 964098, fax +39 089 968739

Marco Pepe Department of Civil Engineering, University of Salerno via Giovanni Paolo II 132, 84084, Fisciano (SA), Italy e.mail: [email protected]

Roberto Realfonzo Department of Civil Engineering, University of Salerno via Giovanni Paolo II 132, 84084, Fisciano (SA), Italy e.mail: [email protected]

Corresponding author: [email protected]

1

ABSTRACT The construction sector is strongly committed to enhancing sustainability of concrete production and several solutions are nowadays under investigation for reducing its environmental impact. This study reports the results of a wide experimental campaign intended at investigating the mechanical and durability performance of structural concretes made with Recycled Concrete Aggregates (RCAs) and coal Fly Ash (FA). To this end, starting from a reference concrete composition, twelve mixtures were designed by replacing part of the ordinary constituents (i.e. cement, sand and coarse aggregates) with the FA and RCAs. The time evolution of the compressive strength, as well as the splitting strength, were measured for monitoring the mechanical performance, whereas the durability performance was scrutinised by measuring water permeability, the carbonation depth and chloride-ions ingress. The obtained results unveil the influence of both RCAs and FA on the resulting concrete performance and highlight that the combined use of these recycled components can lead to a synergistic effect in terms of the relevant physical and mechanical properties of structural concrete.

KEYWORDS: Recycled Aggregate Concrete; Fly Ash; Durability; Water permeability; Chlorideion penetration; Carbonation.

2

1.

INTRODUCTION With a global production of around 10.000 million tons per year [1], concrete is certainly the

most used construction material. Consequently, the concrete production requests huge amounts of energy and raw materials, both resulting in a significant depletion of natural resources [2]. Moreover, mainly due to the cement production processes, it is also responsible for a significant share of the global greenhouse gas emissions [3]. For these reasons, several worldwide organisations and, consequently, different research groups working in the field of construction are nowadays looking for more sustainable solutions intended at the “greening” the concrete industry [4]. A possible solution for pursuing this objective is based on adopting alternative binders (often obtained from industrial by-products) in partial replacement of Portland cement [5], with possible positive consequences not only in terms of sustainability, but also in terms of durability [6]. Otherwise, a further solution is based on replacing ordinary aggregates with recycled ones recycled aggregates [7]. Therefore, this study is manly devoted at investigating the physical and mechanical behaviour of concrete partly made by replacing the ordinary constituents with recycled materials and industrial by-products. Particularly, it reports the results of a wide experimental campaign carried out on samples of thirteen concrete mixtures produced with different combinations of recycled aggregates derived from old existing concrete elements (generally referred to as “Recycled Concrete Aggregates”, RCAs) and coal Fly Ashes (FAs). For the sake of clarity, it is worth mentioning that, in the literature, the concrete produced with recycled aggregates is generally referred to as Recycled Aggregate Concrete (RAC) [8]. As a matter of principle, in order to produce sustainable structural concrete, several recent researches demonstrated that a limited use of recycled aggregates derived from Construction and Demolition Waste (with a particular emphasis on the RCAs) can be a viable solution for enhancing 3

the concrete with negligible consequences on the mechanical performance required in structural applications in terms of technological aspects [9], fundamental behavior (i.e. in terms of cement reaction development) [10][11] and resulting stress-strain response [12]. Nevertheless, since RCAs can be seen as the combination of two main phases, such as the old natural aggregates and the paste attached to them (usually referred to as Attached Mortar, AM [13]), they are generally characterised by higher porosity and water absorption capacity with respect to the “ordinary” natural aggregates. This is a key reason for developing a dedicated mix-design procedure capable of predicting the effects of RCAs on the resulting properties of RAC [14]. Moreover, the rheological behaviour at the fresh state of concrete mixtures can be seriously affected by the amount of RCAs; moreover, the presence of porous media inside concrete can also influence its resulting performance and, hence, recent studies have been devoted to understanding how to control and enhance durability of RAC [15]. In order to mitigate the drawbacks induced by using RCAs in structural concrete and, at same time, promoting a further greening practice in the concrete industry, the use of coal Fly Ash (FA) in (partial) replacement of Portland cement and fine aggregate can also be considered [16]. As widely demonstrated in the literature, the limited use of FA, either as a filler or as a possible cement replacement, can have beneficial effects on RAC in terms of early age properties (workability) [17], mechanical performance [18][19] and long term behaviour [20]. As regards the latter effect, experimental results were recently published that demonstrate the positive effects of FA in terms of chloride-ion penetration resistance [21], water permeability [22] and durability against the exposure to chemical agents and adverse environmental conditions [23]. This paper is mainly devoted at unveiling the consequences of using both RCAs and FA in concrete. The results of a wide experimental campaign, performed at the Materials and Structures Testing Laboratory of the University of Salerno (Italy) and at the Laboratory of the General Admixtures SpA (www.gageneral.com), are reported herein. Particularly, thirteen concrete mixtures 4

were produced by considering the possible combinations of the two following parameters: RCAs replacement ratio and total amount of FA (both used as partial Portland cement and sand replacement). The experimental campaign was aimed at investigating the mechanical and physical properties of the produced mixtures such as the water permeability, the chloride penetration and the carbonation resistance as indexes of their durability performance . Finally, it is worth highlighting that, since the performance of the concrete mixtures under investigation were already analysed in terms of workability and compressive strength [24], the present study rather focuses on the durability-related aspects. The short review of the current state of knowledge, proposed in the first part of this section and based on relevant studies published in very recent time, demonstrates that investigating the durability performance of RAC with FA is indeed a subject of current interest. In this framework, the present paper is characterised by the following aspects of novelty and originality: -

the number of mixtures considered in the present study and the experimental tests carried out for each mixture, also considering the ones already reported in a previous paper [24], represent a wide set of experimental results that may lead to further advances in understanding the potential of combining RCAs and FA and its consequences on both early-age and long-term performance;

-

aggregates replacement (up to a complete substitution of both coarse and fine ones) is considered, along with the use of fly-ash, employed both as a partial sand- and cementreplacement: this makes the variation of relevant parameters significantly wide and, hence, the reported results allow to reach a more comprehensive knowledge about the synergistic effects of RCAs and FA in “green” structural concrete;

-

the results obtained in this study provide information about the possibility of extending the limitations currently provided by the European Code for using this industrial by-product (with and/or without RCAs) as a filler and alternative binder. 5

2.

EXPERIMENTAL PROGRAMME This section outlines materials and techniques employed in this study for producing and

testing the concrete mixtures under investigation. 2.1. Materials The concrete mixtures under investigation were produced by using Portland cement, labelled CEM I 42.4R in accordance with the EN 197-1 [25]. This kind of Portland cement is basically made of pure clinker (i.e., more than 95%) without any addition; it has a specific weight of 3110 kg/m 3. Moreover, class F coal FAs [26] were used for partially replacing both Portland cement and natural sand. The FAs employed herein were produced in a thermo-electrical power plant located in Italy and were compliant with the EN 450-1 [27]. They present a specific weight equal to 2100 kg/m3. Table 1 summarises the chemical composition and physical properties of both the Portland cement and the Fly Ash used in the present study. As regards the aggregates, both Natural and Recycled Concrete Aggregates were used in this study. Crushed limestone particles were employed as natural aggregates, whereas the RCAs were supplied by an Italian company certified according to EN 13242 [28]. These RCAs were obtained by crushing the rubbles derived from demolition of existing concrete structures. Table 2 summarises the key results of the qualification tests carried out on aggregates. Both recycled and natural aggregates were selected, cleaned and sieved in laboratory. Their main physical properties, such as particle density and water absorption capacity at 24 hours, were determined according to the ASTM C127 [29] and ASTM C128 [30], for coarse and fine aggregates, respectively. Natural and recycled concrete aggregates were sorted in four size classes (see Figure 1): -

Class N3, characterised by a nominal diameter of the aggregate particles ranging between 20 mm and 31.5 mm; 6

-

Class N2, characterised by a nominal diameter of the aggregate particles ranging between 10 mm and 20 mm;

-

Class N1, characterised by a nominal diameter of the aggregate particles ranging between 2 mm and 10 mm;

-

Sand, characterised by a nominal diameter of the aggregate particles smaller than 2 mm. Table 3 summarises the results obtained in terms of both particle density and water absorption

capacity for each size class. The results highlight that the RCAs are characterised by lower particle density and the higher water absorption capacity in comparison with the natural ones. This peculiarity is a well-known feature of RCAs [13]. Finally, a chemical admixture based on polycarboxylic polymers, labelled “PRiMIUM RM28” and produced by General Admixtures SpA, was used for controlling the concrete workability at fresh state. 2.2. Mix design and specimens geometry Aimed at evaluating the structural behaviour of the “green” concrete under investigation, thirteen different concrete batches were produced: the first mixture, assumed as a reference, was realised by using only primary aggregates and Portland cement; the other twelve ones were obtained by partly replacing aggregates and Portland cements with RCAs and FA. Table 4 reports the complete composition of the aforementioned mixtures . The first column of the table reports a label denoting the mixture; in particular: -

letters L, M and H indicate the content of fly ash: L stands for low content (i.e. 80 kg/m3); M for medium-to-high content (220 kg/m3); and H for high content (255 kg/m3);

-

letter N indicates those concretes made of only with natural aggregates;

-

letter R indicates that a given percentage of the aggregates was replaced by RCAs; the mentioned percentage is provided through a number following R, i.e. 30, 60 or 100. 7

For example, the mixture LR30 contains the lowest content of fly ash among those considered (i.e. 80 kg/m3) and was obtained by replacing the 30% of natural aggregates with an equivalent volume amount of recycled ones. Other information reported in Table 4 are: the mass per unit volume of cement and of fly ash; the percentage of recycled aggregates used in substitution of natural ones; the water contents per unit volume; three water-binder ratios (their means are explained in the following). As regards the water content, the nominal amount of free water was kept constant (i.e. wfree=150 kg/m3). Then, an “extra” quantity of water (wadd) was added to the various mixtures depending on the different water absorption capacities (at 24 hours) of the aggregates (see Table 3). The wadd amounts, estimated by assuming that aggregates were dried and considering the actual amount of NA and RCA for each sieve fraction, are reported in the fifth column of Table 4. Data related to the control mixture are reported in the first row of Table 4. This a mixture is denoted by the label N and meets the requirements of the EN 206-1 [31] for the exposure class XC2: for this class the maximum water-cement (w/c) ratio is 0.60, the minimum cement content (Cmin) is 280 kg/m3, and the minimum required compressive strength class is C25/30. When FA is added to the mixture, the minimum cement content Cmin may be reduced not more than [31]:





ΔC  C min  200  ρ

(1)

while the maximum amount of FA that can be used as supplementary cementing material is: FA  0.33 by mass  . C

(2)

Moreover, the following relationship has to be satisfied:





C  FA  C min  ΔC  FA  C min

(3)

Coefficient  in eq. (1) is a sort of “cementing efficiency index” of fly ash and in the case of concrete containing CEM-I 42.5  is equal to 0.4 [27]. 8

According to the above assumptions and regulations, for this study, the minimum amount of cement Cmin can be reduced down to 250 kg/m3, and the corresponding maximum amount of FA that can be considered for the partial Portland cement replacement is around 80 kg/m3. In fact, according the EN [27] the extra amount of FA must be considered as “inert” filler for sand replacement. As a matter of fact, the content of fly ash that exceeds the amount given by equation (2) must not take into account for the calculation of the water-binder ratio (Table 4):

w free w free  b C  ρ  min FA; 0.33  C min 

(4)

The following consideration can be drawn by observing the cement and FA contents reported in Table 4: -

the four mixes identified by L as first letter of the label are fully in agreement with the code

provisions: in fact, according to equation (1) the cement content was reduced to C=250 kg/m3 and, correspondingly, 80 kg/m3 of FA were added; -

the group M of four mixes were produced by using 220 kg/m3 of FA and 250 kg/m3 of

cement: therefore, these mixes contain about 140 kg/m3 of FA that cannot be considered as binder and the content of fine aggregates (i.e., the sand) was adjusted for volume balance; -

the third group of four mixes (first letter H) were obtained by further reducing the content of

cement up to 200 kg/m3, i.e. well beyond the limits allowed by equation (1): in order to keep the total volume of binder equal to the one of the second group of mixes 255 kg/m3 of coal FA was used; -

not all the mixes comply with the limit value of the water-binder ratio given by the EN 206-1

Standard for the exposure class XC2 (see last column of Table 4). Concrete mixtures were prepared and casted at the Laboratory of Materials testing and Structures (LMS) of the University of Salerno, Italy. 9

Finally, slump tests were performed on each produced mix in order to control the workability and cubic and cylindrical samples were casted in polyurethane moulds. After 36 h, samples were removed from moulds and cured at 20°C with 100% humidity for 28 days. More detailed information about the produced mixtures under investigation are reported in a companion paper [24]. 2.3. Experimental test matrix The whole experimental programme included tests for evaluating both the mechanical and the durability performances of the studied natural and recycled aggregate concretes. Experimental tests intended at estimating the mechanical properties were performed on standard cubic specimens (150 mm edge) and on standard cylindrical ones (150 mm x 300 mm). In particular, for each of the thirteen concrete mixes the following tests were carried out [24]: -

14 compressive strength tests on cubic specimens at 2, 7, 28, 60, 90 and 365 days of curing (2 tests at 2 days, 1 at 7 days, 6 at 28 days, 2 at 60 days, 1 at 90 days and 2 at 365 days);

-

2 splitting strength tests on cylindrical samples at 28 days, for evaluating the concrete tensile strength.

-

moreover, four standard cubes and one 102 mm x 300 mm cylindrical sample were prepared per concrete mixture in order to perform durability tests. All concrete specimens subjected to durability tests were cured for 28 days at controlled

temperature and humidity; subsequently, they were kept at environmental conditions. For each mixture under investigation the following tests were performed at 90 and 365 days: -

1+1 tests on cubic specimens for estimating the water penetration depth;

-

1+1 always on cubes for evaluating the carbonation depth;

-

2+2 chloride-ion penetration tests on 51 mm x 102 mm cylinders obtained by cutting the

central portion of the above-mentioned 102 mm x 300 mm cylinders into four parts (see Figure 2). Therefore, 104 durability tests were performed in total: 26 water permeability tests, 26 10

carbonation tests and 52 rapid chloride-ion penetrability tests.

11

3.

TESTING PROCEDURES The performance of composite materials is controlled by both the properties of its constituents

and the interactions between them. In concrete, aggregate and cement paste, as well as the interfacial zone between them, affect both the mechanical behaviour and the resulting permeability [32]. Moreover, concrete durability is significantly affected by the latter: as it is well known, concretes with lower permeability have a higher resistance against chemical attacks and are also more capable to resist freeze-thaw cycles [32]. However, concrete permeability is controlled by the so-called "open" porosity and microcracks possibly developed within the concrete matrix may significantly contribute to increasing permeability. Hence, concrete permeability depends on the porosity of the paste, on the aggregate and, consequently, on the paste to aggregate ratio, that define the interface porosity [33]. The whole experimental programme summarised in the following was aimed at investigating the mechanical and physical properties of hardened samples of the aforementioned concrete mixtures made with a variable amount and fraction of natural and recycled aggregates. However, this paper focuses on the durability performance, as the relevant results obtained in terms of mechanical properties were already reported in a previous paper [24]. 3.1. Specific Permeability The “specific permeability” is defined as the rate of flow of a liquid or gas through a porous material under a given pressure, and its value is uniquely function of the pore structure of the considered material [32]. In the case of body subjected to a sufficiently slow unidirectional steady flow, the permeability k can be derived by the following linear law (i.e. the Darcy’s law):

Q

k  A ΔP  μ L

(5)

in which Q is the volumetric flow rate (or “discharge”), µ is the dynamic viscosity of the fluid, A is 12

the nominal cross-sectional of the sample, and DP is the hydrostatic pressure drop over a flow path of length L. According to equation (5) the coefficient of permeability is given by: k

μQ L  A ΔP

(6)

measured in "darcy" (1 darcy = 0.987 µm2). Therefore, a porous system has permeability equal to 1 darcy if a pressure difference of 1 atmosphere (1 atm = 0.101325 × 10-5 Pa) will produce a discharge of 1 cm3/s of a fluid having viscosity equal to 1 centipoise (1 cP = mPa×s) through a sample of length equal to 1 cm. 3.2. Permeability tests The water penetration test, which is nowadays the most commonly used procedure for evaluating permeability in concrete, may be conducted according to the European Standard EN 12390-8 [34]. It consists in measuring the so-called Water Penetration Depth (WPD, in the following) in standard cubic concrete specimens when they are under a water flow due to a given pressure gradient. In particular, water is allowed to penetrate the concrete sample from one side only, by applying a pressure of 0.5 MPa. The water pressure is maintained constant for a period of 72 h. At the end of the test, each specimen is split into two halves for determining the WPD. Then, the profile of the wet portion of the concrete sample can be marked and the maximum value of the WPD can be recorded and considered as an indicator of the water permeability. Twenty-six permeability tests were performed at 90 and 365 days (13+13 tests) according to aforementioned procedure: Figure 3 shows the experimental setup and one specimen after test. The average depth of the water penetration, hm, was evaluated as the ratio between the area of the wet surface and the width of the cubic sample: of course, higher values of hm correspond to higher permeability and, hence, concretes that are supposed to be more easily affected by degradation phenomena. 13

3.3. Rapid chloride penetrability tests Chloride-ion penetration is one of the most harmful phenomena affecting durability of reinforced concrete structures. In fact, although chloride ions in concrete do not directly cause significant damage, they do contribute to the corrosion of steel rebars embedded in the RC structures. Moreover, there are interactions between chloride ions and hydration products of the cement, which may contribute to the development of damage under frost conditions. Rapid Chloride Penetration tests (RCPT) measures the amount of electric charge passing through and allows estimating the concrete resistance to chloride-ion penetration. Therefore, RCPT were carried out on concrete cylinders (51 mm in height and 102 mm in diameter) in accordance with ASTM C1202 [35]: twenty-six samples (i.e. two per mixture) were tested at 90 days and other twenty-six at 365 days. According to the aforementioned international standard, the two ends of each concrete specimen were immersed in a sodium chloride solution and in a sodium hydroxide solution, respectively, before exposing the concrete sample to a potential difference of 60 V for 6 hours (see Figure 4). The total charge passed through the concrete specimen was considered for evaluating the chloride permeability of each sample. A qualitative correlation between the electric charge, expressed in coulombs, and the expected sensitivity to chloride diffusion is reported in Figure 5. 3.4. Assessment of carbonation depth Carbonation is a major cause of deterioration in concrete structures: although it is not a detrimental phenomenon for the concrete itself, it is recognised for triggering corrosion in steel reinforcement [36]. The carbonation rate is controlled by the increase in CO2 concentration inside the concrete pores which, in turn, is influenced by the relative humidity of concrete. As a general trend, within 12 days the surface of fresh concrete reacts with CO2 and, then, the process penetrates deeper into the 14

concrete at a rate proportional to the square root of time; after a year carbonation typically affects depths from about 1 mm (concrete of low permeability made with a low w/b ratio) up to 5 mm or more (porous and permeable concrete made characterised by a high w/b) [37]. Although the carbonation is often considered less serious than the chloride-ions penetration, it is much more widespread since it involves carbon dioxide in the air. Furthermore, carbonation coupled with chloride-ion penetration reduces the durability of concretes exposed to marine environment [32]. The most widely used method for assessing carbonation depth consists in spraying concrete broken faces with a phenolphthalein indicator solution [37]. In this study carbonation tests were performed on standard concrete cubes (150 mm in edge): two specimens per mixture were cast and cured at controlled conditions (in water at 20±2°C) for 28 days. Subsequently, the cubes were placed outside at environmental conditions and tested at 90 and 365 days. At the end of each test, specimens were split into two parts and the phenolphthalein solution (the indicator was obtained by dissolving 1% phenolphthalein in isopropyl alcohol) was applied on the broken surfaces. The “affected depths” - i.e. the distances of the coloured parts from the external surface - were readily shown and were easily measured (see Figure 5). Finally, the carbonation depth D was evaluated as follows [38]: D

A1  A 2  B1  B 2  C1  C 2  D1  D 2 8

(7)

15

4.

RESULTS AND ANALYSIS This section reports and analyses the experimental results obtained in this study. Particularly,

it briefly summarises the key results in terms of physical and mechanical characterisation of the sustainable structural concrete mixtures under investigation as they were already extensively discussed by the authors in a previous paper [24]. Then, the analysis focusses on the influence of using recycled components (i.e., RCAs and FAs) on the resulting long-term durability performance of the mixtures under investigation. 4.1. Physical characterization of sustainable structural concrete: mass density As it is well known, the mass density of concrete depends on the amount and density of coarse and fine aggregates, the amount of entrained air (and entrapped air) and on the water content. “Ordinary” structural concrete is generally characterised by a mass density around 2400 kg/m 3 and, as a matter of fact, lower values of this quantity results from higher water contents, which, in turn, lead to lower mechanical strength. Figure 6 reports the mean values of mass densities of the concrete mixtures examined in this study. They were measured by weighting 6 standard cubes (150 mm edge) per concrete mix before performing the compressive strength tests at 28 days of curing age. Figure 6 highlights that lower mean value of the concrete density were obtained for higher values of aggregate replacement ratio: this result was actually expected as RCAs are more porous than NAs [9]. 4.2. Mechanical properties 4.2.1. Time evolution of compressive strength Figure 7 to Figure 9 highlight the time evolution of the compressive strength of the concrete mixtures under investigation. Particularly, they report the experimental data determined at 2, 7, 28, 16

60, 90 and 365 days by reporting the average value measured for each curing age. The analysis of the results plotted in Figure 7 to Figure 9 points out the key roles of both FAs and RCAs on the resulting time evolution of the compressive strength. First of all, the role of the coal Fly Ash is clearly highlighted by these results. As a matter of fact, it is apparent that before 28 days of curing, the concrete strength of the mixtures with FA grows up slower than the ordinary concrete (i.e., the N control mix) and, hence, they do not reach the same 28 day strength exhibited by the control mixture. Conversely, the mixtures including FA show a significant increase in strength and, then, after a sufficiently longer curing time they reach compressive strength values higher than those obtained for the control mix, whose strength is almost constant after the conventional 28 day curing duration. This result is mainly due to the delayed binding action of the fly ash, which needs longer times with respect to those required for the Portland cement [20]. Therefore, the compressive strength obtained for concrete mixtures produced with FAs and natural aggregate only (i.e., mixtures LN, MN and HN) is higher than the reference values and, although in the MN and HN mixtures part of the fly ashes should be considered as an inert filler [27], it actually contributes to the resulting mechanical property at long curing age. In fact, they show a different delayed binding effect that is more pronounced in the case of mixtures with higher amounts of FA. Therefore, it is clear that the limitation proposed by the EN [27] in terms of the effective contribution of FA to the mixture binder (see equation (2)) should be extended in the case of curing ages longer than the conventional 28 days. Moreover, as regards the role of RCAs in concrete, Figure 7 to Figure 9 highlight that the use of RCA in concrete affects its compressive strength. This effect can certainly be attributed to the higher porosity of RCA, as it results in a weaker structure in the solid skeleton of the concrete matrix. Nevertheless, it is worth highlighting that recent studies demonstrated that this decay in strength of RAC is mainly due to the "effective" free water actually available for the cement 17

hydration: it is not generally equal to the "nominal" free water, as the initial moisture conditions and the higher water adsorption capacity of RCAs can significantly affect the amount of water that is available for cement hydration reaction [9]. Since structural concrete mixtures are generally produced by using initially dry aggregates and adding water for their full saturation (considering their water absorption capacity at 24h), when this procedure is applied to RACs, the higher water absorption capacity of RCAs is higher than in ordinary mixtures and, consequently, it significantly modifies the "effective" water-to-cement (or binder) ratio. Pepe [13] suggests to consider that during the mixing time RCAs are capable to absorb 50% of their water absorption capacity at 24 h and, then, the "effective" free was defined by adding the nominal one wfree and 50% the added water wadd (see Table 2). Therefore, the analysis proposed in the following and the conclusions drawn out about the influence of FA and RCA in the time evolution of compressive strength can be confirmed by interpreting the experimental data in the light of this definition of effective water and water-tobinder ratio (w/b)eff: w  0.5  w add w    C  ρ  FA  b  eff

Figure 10 shows that a close correlation emerges between the

(8) (w/b)eff values and the

corresponding compressive strength Rcm, especially for long curing durations. As a matter of fact, the points group around a curve resembling the Abram´s law, widely accepted for expressing the relationship between w/c and Rcm in ordinary concrete. This observation confirm the sound definition of the ratio (w/b)eff, intended at generalising the concept of w/c for taking into account the various peculiarities of the concrete mixtures under consideration, such as the effect of FA, the initial moisture condition and the actual water absorption capacity of coarse aggregates. 4.2.2. Splitting tensile strength The tensile strength of the examined concrete were determined via splitting tests on 18

cylindrical samples at 28 days of curing age [39]: the mean values fctm are depicted in Figure 11. These results demonstrate that, at 28 days of curing, the concretes made with natural aggregates and FA exhibit a tensile strength higher than the one of the control mixture. However, the actual content of FA does not affect significantly the tensile strength of the concretes under consideration. Moreover, as well as in the case of compressive strength, the tensile strengths progressively decrease by increasing the replacement ratio of natural aggregates with recycled ones. 4.3. Durability-related properties 4.3.1. Results of water permeability tests As already mentioned, permeability of the thirteen concrete mixtures under investigation was determined at 90 and 365 days of curing through water injection, according to the testing procedure indicated by the EN 12390 [34]. Figure 12 depicts the contour lines describing of the wet surface resulting at the end of some of these permeability tests. Figure 13 reports the values of average water penetration depth hm (defined in section 3.2) obtained from the permeability tests . Unfortunately, the results of tests performed on cubic samples made of the mixture LR100 are not available, as they failed under the water pressure at both 90 and 365 days: in fact, due to the presence of some microcracks, subsequently enlarged by the action of the water under pressure, they were completely overpassed by the water, from one side to the other. The following considerations can be drawn by observing the data shown in Figure 13: -

almost all the concrete mixtures produced with RCAs and FA show a lower water permeability capacity in comparison with the “ordinary” control mixture N, both at 90 and 365 of curing (the only exception are the mixtures LN and LR60 that, at 90 days, show almost the same permeability of the reference mixture);

-

except for the reference mixture, the difference in water penetration measured on specimens cured for 90 and 365 days are in the order of magnitude of the dispersion generally affecting 19

experimental observations of the physical properties in concrete; -

the concrete permeability decreases significantly with increasing the content of FA;

-

the concrete permeability appears not to be significantly influenced by the total amount of RCAs. The role of FA clearly emerges by these experimental observations. It influences the

microstructure of paste and enhances the water tightness of the concrete, also counterbalancing the expected increase in porosity and, hence, water permeability due the presence of RCAs. As a matter of principle, the lower mean value of the nominal diameter characterising the grain size distribution of fly ash compared with cement particles, leads to a reduction of voids inside the concrete microstructure and, hence, it directly influences the water permeability more than the aggregate replacement. Therefore, concrete mixes containing the highest amounts of FA are those characterise by the lowest water permeability. Moreover, as regards RCAs, it is well known that they are generally more porous than natural aggregates [13] but nevertheless, despite some studies have demonstrated that the concrete permeability increases by substituting amounts of NAs with RCAs, Figure 14 show that in this research the influence of recycled aggregates is negligible. In fact, Figure 14 reports the average value of water permeability measured, both at 90 and 365 days, for each group mixtures characterised by the same total amount of FA (i.e., L, M and H). Moreover, the error lines reported in the bars represent the scatter of results between the mixtures characterised by the same total amount of binders, but different amount of RCAs. Since the error bars do not show significant variations, it is possible to confirm that the resulting water permeability is significantly more affected by the content of FA (that, in fact, results in a more compact concrete paste) than the one of RCAs. 4.3.2. Results of rapid chloride-ion penetrability tests Fifty-two rapid chloride-ion penetrability tests introduced in section 3.3 were performed on 20

concrete cylinders at both 90 and 365 of curing (that is to say, four tests per concrete mixture and twenty-six per curing age). Figure 15 reports the electric charge (measured in Coulombs, C) passing through each cylinder, from side to side, in 6 hours. Since the greatest resistances to chloride-ion penetration correspond to lower values of the electric charge, the results shown in Figure 15 demonstrate that a significant attenuation of the chloride-ion penetration phenomenon can be achieved by adding FA to the concrete mixtures. Conversely, the higher the percentage of RCAs, the lower the resistance to chloride-ion penetration. Furthermore, all the concrete mixtures containing FA show resistances to the chloride-ion penetration at 365 days remarkably lower than those achieved at 90 days. This can be attributed to the aforementioned delayed reaction of FA that contributes to make the concrete mortar less permeable to chlorides. Finally, the obtained experimental results clearly indicate that the water permeability data cannot be used to predict the concrete resistance to the chloride-ion penetration and vice versa, especially when RCAs are used in concrete. As already evidenced in the technical literature, this discrepancy may be related to the different transport mechanisms involved in the water permeability test and in the chloride penetration tests [33]. 4.3.3. Results of carbonation tests Two standard cubic samples (150 mm edge) per concrete mix were tested at 90 and 365 days of curing in order to assess the carbonation depth as already described in section 3.4. The pictures collected in Figure 16 to Figure 18 highlight the resulting carbonation depth D of the tested samples. Particularly, they show the cross-sections of specimens made with low, medium and high content of FA, respectively. Figure 19 depicts the obtained carbonation depths, determined by means of equation (7). It is worth highlighting that the large variability in the constituents of the thirteen mixtures (i.e. percentage of RCAs; amount of FA; water content-cement ratio; amount of Portland cement) does 21

not allow for easily identifying the influence of some of them in influencing the carbonation level. As expected Figure 19 confirms that carbonation is higher for concrete specimens tested after 365 than 90 days of curing. All the mixtures realised with RCAs and FAs were generally affected by a higher value of carbonation depth with respect to the control mixture N. In addition, the presence of RCAs seems to play a role in the carbonation resistance. Since carbonation is mainly controlled by the chemical, physical and mechanical characteristics of the paste present in concrete, it would be useful to analyse the experimental data registered for the carbonation depth D by considering the effective water-to-binder ratio defined by equation (8). As a matter of principle, this parameter can indirectly indicate the key characteristics of the paste of concrete. In fact, Figure 20 highlights the influence of (w/b)eff on the resulting carbonation depth: as expected, higher values of the water/binder ratio result in deeper carbonation layers. This analysis highlights that, as in the case of water permeability, the carbonation is more influenced by the paste than the aggregates. In fact, the mixtures with high percentage of RCAs show higher values of D, as they are also characterised by the higher water/binder ratios. Moreover, as well know from the literature, the carbonation depth D can be correlated with time through the following equation [40]: D  kc  t

(9)

where kc is the carbonation factor and t is the air exposure time (expressed in days). The carbonation factor was determined for all the produced mixtures and, for instance, Figure 21 reports the calibration determined for some of the employed mixtures, whereas Table 6 reports the carbonation factor calculated for each mixture. Finally, the emerging correlation between the carbonation factor and the amount of FA and RCA confirms the above statement (Figure 22).

22

5.

CONCLUSIONS

This paper analysed the mechanical and durability-related performance of sustainable structural concrete produced with recycled concrete aggregates and fly ash. The following conclusion can be drawn out: -

the results presented herein demonstrated the feasibility of using recycled constituents for

producing sustainable concrete mixtures for structural applications: as a matter of fact, both in term of mechanical and durability performance, the mixture produced with even high amount of RCAs and FAs presents acceptable results for structural application requirements; -

the analysis of the results and the introduction of an effective water-to-binder ratio highlighted

that the weaker mechanical and durability behaviour of RACs, generally induced by the high porosity of the recycled aggregates, can be correlated with the higher amount of water/binder ratio that derive from the standard procedures adopted for structural concrete production; -

the water permeability of concrete is mainly governed by the paste characteristics than the

aggregate ones; moreover, the presence of FA in concrete reduces the permeability capacity of the composite and, in addition, the presence of high volume of FA mitigates the possible detrimental effects due to the use of RCAs; -

the chloride ions penetration is affected by the presence of RCA in concrete, but, at the same

time, the use of FA enhances the resistance to chloride-ion penetration; -

the carbonation depth is mainly controlled by the paste characteristics: it is higher for higher

values of the effective water-to-binder ratio defined in this study; -

as a general comment, the results obtained in this study demonstrate that the limitations

currently provided by EN 206 [31] for using FA as a concrete constituent may be further extended 23

without loss of performance, also in combination with RCAs. Finally, the results reported and analysed herein, along with the other ones recently published in the international scientific literature, can contribute at assembling a wide database for future modelling of the mechanical and durability-related phenomena occurring when the considered recycled constituents and industrial by-products are employed in structural concrete.

24

ACKNOWLEDGEMENTS This study is part of the activities carried out by the authors within the ‘‘EnCoRe’’ project (www.encore-fp7.unisa.it), funded by the European Union within the Seventh Framework Programme (FP7-PEOPLE-2011-IRSES, n. 295283). The authors are grateful to General Admixtures S.p.A. (www.gageneral.com) for its important support to the experimental investigation.

25

REFERENCES [1]. Hasanbeigi A, Price L, Lin E. Emerging energy-efficiency and CO 2 emission-reduction technologies for cement and concrete production: A technical review. Renewable and Sustainable Energy Reviews 2012; 16(8): 6220-6238. [2]. Meyer C. The greening of the concrete industry. Cement and Concrete Composites 2009; 31(8): 601-605. [3]. Glavind M. Sustainability of cement, concrete and cement replacement materials in construction. Khatib editor, Sustainability of Construction Materials, Wood Head Publishing in Materials 2009; 120–147. [4]. Pacheco-Torgal F, Jalali S, Labrincha J, John VM. Eco-efficient concrete. Elsevier 2013. [5]. Juenger MCG, Winnefeld F, Provis JL, Ideker JH. Advances in alternative cementitious binders. Cement and Concrete Research 2011; 41(12): 1232-1243. [6]. Aïtcin PC. Binders for durable and sustainable concrete. CRC Press 2007. [7]. de Brito J, Saikia N. Recycled aggregate in concrete: use of industrial, construction and demolition waste. Springer Science & Business Media 2012. [8]. Behera M, Bhattacharyya SK, Minocha AK, Deoliya R, Maiti S. Recycled aggregate from C&D waste & its use in concrete–A breakthrough towards sustainability in construction sector: A review. Construction and building materials 2014; 68: 501-516. [9]. Marinković SB, Ignjatović IS, Radonjanin VS, Malešev MM. Recycled aggregate concrete for structural use–an overview of technologies, properties and applications. Innovative Materials and Techniques in Concrete Construction 2012; 115-130. [10]. Koenders EAB, Pepe M, Martinelli E. Compressive strength and hydration processes of concrete with recycled aggregates. Cement and Concrete Research 2014; 56: 203-212. [11]. Carneiro JA, Lima PRL, Leite MB, Toledo Filho RD. Compressive stress–strain behavior of steel fiber reinforced-recycled aggregate concrete. Cement and Concrete Composites 2014; 46: 65-72. [12]. Thomas C, Setién J, Polanco JA, Alaejos P, Juan MSD. Durability of recycled aggregate concrete. Construction and Building Materials 2013; 40: 1054-1065. [13]. de Juan MS, Gutiérrez PA. Study on the influence of attached mortar content on the properties of recycled concrete aggregate. Construction and Building Materials 2009; 23(2): 872-877. [14]. Pepe M. A conceptual model to design Recycled Aggregate Concrete for structural applications. Ph.D. Thesis, Department of Civil Engineering, University of Salerno, 2015. [15]. Kou SC, Poon CS. Enhancing the durability properties of concrete prepared with coarse recycled aggregate. Construction and Building Materials 2012; 35: 69-76. 26

[16]. Corinaldesi V, Moriconi G. Influence of mineral additions on the performance of 100% recycled aggregate concrete. Construction and Building Materials 2009, 23(8): 2869-2876. [17]. Kim K, Shin M, Cha S. Combined effects of recycled aggregate and fly ash towards concrete sustainability. Construction and Building Materials 2013, 48: 499-507. [18]. Lotfy A, Al-Fayez M. Performance evaluation of structural concrete using controlled quality coarse and fine recycled concrete aggregate. Cement and Concrete Composites 2015, 61: 3643. [19]. Limbachiya M, Meddah MS, Ouchagour Y. Use of recycled concrete aggregate in fly-ash concrete. Construction and Building Materials 2012, 27(1): 439-449. [20]. Kou SC, Poon CS. Long-term mechanical and durability properties of recycled aggregate concrete prepared with the incorporation of fly ash. Cement and Concrete Composites 2013; 37: 12-19. [21]. Sim J, Park C. Compressive strength and resistance to chloride ion penetration and carbonation of recycled aggregate concrete with varying amount of fly ash and fine recycled aggregate. Waste management 2011, 31(11): 2352-2360. [22]. Zong L, Fei Z, Zhang S. Permeability of recycled aggregate concrete containing fly ash and clay brick waste. Journal of Cleaner Production 2014, 70: 175-182. [23]. Shin M, Kim K, Gwon SW, Cha S. Durability of sustainable sulfur concrete with fly ash and recycled aggregate against chemical and weathering environments. Construction and Building Materials 2014, 69: 167-176. [24]. Lima C, Caggiano A, Faella C, Martinelli E, Pepe M, Realfonzo R. Physical properties and mechanical behaviour of concrete made with recycled aggregates and fly ash. Construction and Building Materials 2013, 47: 547-559. [25]. EN 197-1. Cement Composition, specifications and conformity criteria for common cements. 2011. [26]. ASTM C618. Standard Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use in Concrete. 2012. [27]. EN 450-1. Fly ash for concrete Definition, specifications and conformity criteria. 2012. [28]. EN 13242. Aggregates for unbound and hydraulically bound materials for use in civil engineering work and road construction. 2013. [29]. ASTM C127. Standard Test Method for Density, Relative Density (Specific Gravity), and Absorption of Coarse Aggregate. 2012. [30]. ASTM C128. Standard Test Method for Density, Relative Density (Specific Gravity), and Absorption of Fine Aggregate. 2012. [31]. EN 206-1. Concrete Specification, performance, production and conformity. 2000. [32]. Mather, B. Concrete durability. Cement and Concrete Composites 2004; 26(1): 3-4. 27

[33]. Chia KS, Zhang MH. Water permeability and chloride penetrability of high-strength lightweight aggregate concrete. Cement and Concrete Research 2002; 32: 639 – 645. [34]. EN 12390-8. Testing hardened concrete - Part 8: Depth of penetration of water under pressure. 2009. [35]. ASTM C1202. Standard Test Method for Electrical Indication of Concrete's Ability to Resist Chloride Ion Penetration. 2012. [36]. Bertolini L, Elsener B, Pedeferri P, Redaelli E, Polder R.B. Corrosion of steel in concrete: prevention, diagnosis, repair. John Wiley & Sons 2013. [37]. Pacheco Torgal F, Miraldo S, Labrincha JA, De Brito J. An overview on concrete carbonation in the context of eco-efficient construction: Evaluation, use of SCMs and/or RAC. Construction and Building Materials 2012; 36: 141–150. [38]. EN 14630. Products and systems for the protection and repair of concrete structures. Test methods. Determination of carbonation depth in hardened concrete by the phenolphthalein method. 2006. [39]. EN 12390-6. Testing hardened concrete - Part 6: Depth Tensile splitting strength of test specimens. 2009 [40]. Silva RV, Neves R, de Brito J, Dhir RK. Carbonation behaviour of recycled aggregate concrete. Cement and Concrete Composites 2015; 62: 22-32.

28

Natural Sand

Natural N1

1 cm

Recycled Sand

1 cm

1 cm

1 cm

1 cm

Recycled N3

Recycled N2

Recycled N1

1 cm

Natural N3

Natural N2

1 cm

1 cm

Figure 1: Detail of natural and recycled concrete aggregates.

29

Figure 2: Concrete cylinders used for chloride-ion penetration tests.

30

Wet area = 5483.00 mm2 Dimension of the cube = 150.00 mm hm =

36.55 mm

Figure 3: Water permeability tests.

31

Figure 4: Chloride-ion penetration tests.

32

Figure 5: Carbonation test.

33

2600 control mixture "N"

density [kg/m3]

2400

FA content

group "L"

2200

group "M" 2000

group "H" 1800 1600 0

30

60

100

% of RCAs Figure 6: Density of concrete mixtures.

34

Rcm [MPa]

70

N

60

50

LN

40 LR30

30 LR60

20 10

LR100

0 2

7

28

60

90

365

age [days] Figure 7: Compressive strength of NAC and RAC with low content of FA.

35

Rcm [MPa]

70 N

60 50

MN

40

MR30 30 MR60

20

10

MR100

0 2

7

28

60

90

365

age [days] Figure 8: Compressive strength of NAC and RAC with medium content of FA.

36

Rcm [MPa]

70

60

N

50

HN

40 HR30

30

HR60

20 10

HR100

0

2

7

28

60

90

365

age [days] Figure 9: Compressive strength of NAC and RAC with high content of FA.

37

Rcm [MPa]

Rcm [MPa]

70 28 days

60 50

40 30 20

70 60 days

60 50 40

N

30

L

20

M 10

0.40

0.50 0.60 (w/b)eff

0.70

70 90 days

60 50

40

H

0 0.30

0.80

Rcm [MPa]

Rcm [MPa]

L M

10

H

0 0.30

N

0.40

0.80

365 days

60 50 40

N

30

N

20

L

20

L

M

0 0.30

0.70

70

30

10

0.50 0.60 (w/b)eff

M 10

H 0.40

0.50 0.60 (w/b)eff

0.70

0.80

0 0.30

H 0.40

0.50 0.60 (w/b)eff

0.70

0.80

Figure 10: Compressive strength vs effective water-to-binder ratio.

38

fctm [MPa]

4.0 3.5 3.0

control mixture "N"

FA content

2.5

group "L"

2.0

group "M"

1.5

group "H"

1.0 0.5 0.0

0

30 60 % of RCAs

100

Figure 11: Splitting tensile strength of concrete mixtures at 28 days.

39

Figure 12: Some results from water permeability tests.

40

80 70 f a i l u r e

hm [mm]

60

50 40 30 20 10 0

N 90 days [mm] 36.55 365 days [mm] 75.45

LN 37.00 57.09

LR30 32.34 40.71

LR60 43.11 38.82

LR100 -

MN 18.69 15.89

MR30 16.17 19.55

MR60 MR100 14.69 21.40 16.63 26.55

HN 24.76 23.32

HR30 14.79 16.98

HR60 10.09 17.25

HR100 10.56 10.83

Figure 13: Experimental values of the water penetration depth tests.

41

100 90 days 80

hm [mm]

365 days 60 40 20 0 Gruop "L"

Gruop "M"

Gruop "H"

Figure 14: The infleuce of RCAs and FAs on water penetration depth for concrete.

42

1.200

8000 11900

1.400

2571

passing electric charge [C]

1.600 low

1.000 800 very low

600 400 200 0 90 days 365 days

negligible N 1114 2571

LN 389 137

LR30 790 364

LR60 907 560

LR100 8000 11900

MN 93 34

MR30 207 50

MR60 317 125

MR100 1330 347

HN 221 47

HR30 240 98

HR60 326 165

HR100 801 359

Figure 15: Results of chloride ion penetration tests.

43

90 days 365 days Figure 16: natural carbonation at 90 and 365 days of concrete samples containing low amounts of FA.

44

90 days 365 days Figure 17: natural carbonation at 90 and 365 days of concrete samples containing medium amounts of FA.

45

90 days 365 days Figure 18: Natural carbonation at 90 and 365 days of concrete samples containing highest amounts of FA.

46

D [mm]

10 9 8 7 6 5 4 3 2 1 0 90 days 365 days

N 1.1 1.5

LN 2.4 4.1

LR30 2.5 5.6

LR60 3.5 7.0

LR100 6.0 7.9

MN 1.5 1.9

MR30 2.4 2.9

MR60 MR100 3.1 3.8 4.1 4.4

HN 3.5 5.5

HR30 3.8 5.8

HR60 4.5 6.5

HR100 5.6 8.7

Figure 19: Carbonation depths of all concrete specimens.

47

9 8 7 6 5 4 3 2 1 0 0.30

Carbonation depth - D [mm]

Carbonation depth - D [mm]

10 90 days

N L

M H 0.40

0.50 0.60 (w/b)eff

0.70

0.80

10 9 8 7 6 5 4 3 2 1 0 0.30

365 days

N L

M H

0.40

0.50 0.60 (w/b)eff

0.70

0.80

Figure 20: The influence of the effective water-to-binder ratio on carbonation depth for sustainable concrete.

48

Carbonation D [mm]

7 N

6

L 5

M

4

H

3 2 1 0 0

0.2

0.4

0.6

0.8

1

year1/2 Figure 21: Calibration of the carbonation coefficient for the concrete mixtures.

49

Carbonation coefficient - kC

10 9 8 7 6 5 4 3 2 1 0

N 0

20

40 60 RCA [%]

L

M 80

H 100

Figure 22: Influence of RCAs and FAs on the carbonation coefficient for concrete.

50

Table 1: Chemical composition and physical properties of cement and fly ash. Chemical compound

CEM I 42.5 R

Fly ash

CaO (%)

64,06

2,30

SiO2 (%)

18,90

46,90

Al2O3 (%)

4,90

28,50

Fe2O3 (%)

3,66

6,22

SO3 (%)

2,92

0,04

MgO (%)

0,82

1,23

Loss to Ignition (%)

3,18

6,20

Specific gravity (kg/m3)

3110

2100

51

Table 2: Sieved aggregates used in concrete mixtures. Index

RCA

NA SI 15

Shape Index

EN 933-4

SI 20

Flakiness Index

EN 933-3

FI 20

FI 15 3

Particle density

EN 1097-6

2369 kg/m

2690 kg/m3

Water absorption

EN 1097-6

1.8 – 12.2 %

0.3 – 1.2 %

52

Table 3: Size fractions, mass densities and water absorption capacity of NAs and RCAs. Fraction range [mm] N3:

20 - 31.5

N2:

10 – 20

N1:

2 – 10

Sand:

0-2

Mass density [kg/m3] Natural Recycled Aggregates Aggregates

2690

2370

Water absorption [%] Natural Recycled Aggregates Aggregates 0.3 1.8 0.5

3.0

0.7

6.0

1.2

12.2

53

Table 4: Concrete mixtures. Mixture

CEM I 42.5R 3

N LN LR30 LR60 LR100 MN MR30 MR60 MR100 HN HR30 HR60 HR100

FA

RCA 3

wfree

wadd

3

3

[kg/m ]

[kg/m ]

[%]

[l/m ]

[l/m ]

280 250 250 250 250 250 250 250 250 200 200 200 200

0 80 80 80 80 220 220 220 220 255 255 255 255

0 0 30 60 100 0 30 60 100 0 30 60 100

150 150 150 150 150 150 150 150 150 150 150 150 150

14.02 14.02 21.86 38.16 109.65 11.28 17.85 28.68 98.84 11.28 17.85 28.68 98.84

wfree/b

(w/b)eff

0.54 0.53 0.53 0.53 0.53 0.53 0.53 0.53 0.53 0.64 0.64 0.64 0.64

0.56 0.56 0.57 0.60 0.73 0.46 0.47 0.49 0.59 0.52 0.53 0.54 0.66

54

Table 5: Chloride ions penetrability vs charge passed. Charge Passed (coulombs)

Chloride ions Penetrability

> 4 000

High

2000 – 4 000

Moderate

1000 – 2 000

Low

100 – 1 000

Very Low

< 100

Negligible

55

Table 6: Chloride ions penetrability vs charge passed. Mix N LN LR30 LR60 LR100 MN MR30 MR60 MR100 HN HR30 HR60 HR100

D [mm] 90 days 1.1 2.4 2.5 3.5 6.0 1.5 2.4 3.1 3.8 3.5 3.8 4.5 5.6

365 days 1.5 4.1 5.6 7.0 7.9 1.9 2.9 4.1 4.4 5.5 5.8 6.5 8.7

KC

R2

1.64 4.25 5.49 7.01 8.73 2.22 3.28 4.52 5.04 5.81 6.17 7.01 9.21

0.92 0.99 1.00 1.00 0.90 0.88 0.85 0.90 0.82 0.97 0.96 0.94 0.97

56