Effect of fly ash and superabsorbent polymer on concrete self-healing ability

Construction and Building Materials 233 (2020) 116975

Contents lists available at ScienceDirect

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

Effect of fly ash and superabsorbent polymer on concrete self-healing ability Pattharaphon Chindasiriphan ⇑, Hiroshi Yokota, Paponpat Pimpakan Hokkaido University, Sapporo, Japan

a r t i c l e

i n f o

Article history: Received 9 March 2019 Received in revised form 8 August 2019 Accepted 14 September 2019

Keywords: Crack-healing Self-healing concrete Fly ash Superabsorbent polymer Self-healing materials

a b s t r a c t Concrete that contains self-healing supplementary materials is a feature of many sustainable structures because it decreases maintenance costs and extends service life. Many self-healing studies have suggested that concrete needs to have extensive exposure to water in order to promote crack-sealing. This paper investigates the benefits of using fly ash and superabsorbent polymer (SAP) to promote selfhealing. Eight mix proportions with varying fly ash and SAP replacement ratios were examined. Selfhealing efficiency was evaluated by temporal decreases in water discharge through a crack, crack closure observation, stereomicroscopy, scanning electron microscopy (SEM), thermogravimetric/differential thermal analysis (TG/DTA), and energy-dispersive X-ray spectroscopy (EDS). The specimens were preloaded to generate a crack, and these pre-cracked specimens were healed either by continuous water immersion or exposure to wet-dry conditions. Crack width was found to decrease with increase in the fly ash replacement ratio, whereas higher volume of SAP was found to mitigate water discharge through the crack by the swelling of the SAP gel. These effects in combination achieved a maximum of 100% crack closure and permeability restoration by 28 days of healing. As the results, midterm self-healing ability of concrete can be enhanced by the coupling effect of fly ash and SAP. The crack closure was found to be associated with the development of self-healing products such as calcium carbonate and C-S-H. Ó 2019 Elsevier Ltd. All rights reserved.

1. Introduction Infrastructure is fundamental to socioeconomic activities. The most common material for civil engineering work is concrete, because of its favorable mechanical properties, durability and economy. Concrete structures deteriorate due to shrinkage, excessive loading and exposure to severe environments. Constraints on inspection and maintenance budgets are an inevitable issue for many concrete structures. Cracked concrete is prone to deterioration, as the cracks allow harmful agents to infiltrate and cause steel corrosion. It was reported that around £40 billion is spent annually for structural maintenance in the UK, a significant share of which is used to repair deteriorated concrete structures [1]. This was confirmed by Gardner et al. [2], who reported that about a half of the infrastructure budget in the EU goes to the maintenance, repair and replacement of damaged concrete structures. This evidence suggests that budgetary constraints are a regional issue, with infrastructure construction outpacing maintenance. Proper maintenance is neglected for some structures, thereby jeopardizing their structural soundness, serviceability and safety. ⇑ Corresponding author. E-mail address: [email protected] (P. Chindasiriphan). https://doi.org/10.1016/j.conbuildmat.2019.116975 0950-0618/Ó 2019 Elsevier Ltd. All rights reserved.

‘‘Self-healing concrete” refers to a material that is capable of recovering its original properties after suffering degradation. Autogenous crack-healing may be observed spontaneously when water infiltrates small cracks (less than 0.3 mm wide) [3–9]. Crackhealing efficiency is influenced by many parameters, such as initial crack width, healing duration, hydraulic pressure gradient, healing temperature, mix design, and exposure conditions. Extensive research has found crack closure to be dependent on several mechanisms, such as those incorporating autogenous self-healing by the continued hydration of unreacted cement, calcium carbonate crystallization, the deposition of impurities in water, and the swelling and expansion of C-S-H gel including C-S-H formation from pozzolanic reactions [5,9–11]. The current trend of interdisciplinary approaches to the development of self-healing concrete shows attempts to seek cementitious supplementary materials that afford higher efficiency and greater predictability. ‘‘Autonomous healing” refers to a healing mechanism that is engineered to enhance self-healing [12]. Encapsulation methods are being widely developed, such as those involving brittle capsules or hollow tubes filled with calciteprecipitating bacteria, other microorganisms, polymer fillers, or

2

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

adhesives. The healing process is triggered by concrete cracking, when capsules containing the healing materials are broken [13–15]. Advances in materials engineering have given rise to alternative materials, such as shape-memory polymers and shape-memory alloys integrated with cementitious materials. The rehabilitation process is activated by heat, which allows a material of a predetermined shape to deform into a preset configuration and subsequently generate restraining force to close an open crack [16]. To achieve practical autonomous healing in construction, an approach that uses superabsorbent polymers (SAPs) has been proposed. Lee et al. [17] proposed SAPs as materials that satisfy many criteria, including cost-effectiveness, durability and toughness. SAPs or hydrogels are water-entrained admixtures that are able to absorb great volumes of liquid and that subsequently swell to become watertight impermeable gels. Jensen and Hansen [18] suggested that SAP decreases the possibility of self-desiccation shrinkage by providing extensive internal curing. Riyazi et al. [19] noted that SAP is able to mitigate freeze–thaw damage by achieving a satisfactory air void distribution. In the practical application of self-healing, Lee et al. [17] revealed that SAP swells after exposure to water. SAP expansion enables SAP to fill cracks and other cavities in concrete and to subsequently restore the permeability of the concrete system. Fly ash is an industrial byproduct of pulverized coal combustion. Jonkers et al. [20] noted that the cement industry generates approximately 7% of all global anthropogenic carbon dioxide emissions. Pioneering research replaced cement with fly ash to reduce the proportion of Portland cement, which has allowed cement manufacturers to optimize material costs and environmental friendliness [21,22]. Fly ash contributes to long-term strength development, which results in a greater final compressive strength due to the pozzolanic reaction that occurs at the late stage. Thomas [23] noted that the calcium hydroxide generated by the hydration reaction of Portland cement is consumed by the pozzolanic reaction from fly ash to produce calcium-silicate hydrate (C-S-H) and calcium-aluminate hydrate and that these subsequently improve the compressive strength and reduce the porosity of the concrete matrix. SAP utilizes the water that accumulates at damaged locations [17,24]. As stated earlier, the pozzolanic reaction afforded by fly ash inclusion is known to improve the long-term strength and to reduce the permeability of the concrete. A fly ash–cement system was reported by Termkhajornkit et al. [25] that the progress of pozzolanic reaction in later stage showed a positive impact on crack closure. Zhang et al. [26] concluded that self-healing of microcrack was observed with a partial mechanical properties recovery in the engineered cementitious composites (ECC). Despite the fact that fly ash and SAP are the common materials that have been long used in concrete work because of their versatility, availability, cost effective and sustainability. The coupling effect of fly ash and SAP on medium to long-term self-healing performance and the potential of using them in wet-dry exposure have not yet well comprehended. This study investigates the effect of fly ash and SAP on concrete self-healing capability, considering eight classes of cement mortar under two different exposure conditions, such as continuous water immersion and exposure to wet-dry. Slump testing and compressive strength testing were carried out to investigate a feasible mix design and to determine the behavior of concrete engineered with fly ash and SAP. Self-healing performance was evaluated by an analysis of physical properties that included reduction in the volume of water discharge through a crack and visual crack closure observation. The results of physical properties analyses were used as a statistical reference for self-healing performance assessment by microstructure investigation done by stereomicroscopy and

SEM. In addition, chemical analysis by thermogravimetric/differen tial thermal analysis (TG/DTA) was made to elucidate the effect of carbonation on self-healing, and energy dispersive X-ray spectroscopy (EDS) was used to identify newly formed self-healing materials. 2. Experimental investigation 2.1. Material The cement mortar consisted of ordinary Portland cement (OPC), sand, SAP, and fly ash is a target of this study. Fly ash with high CaO content (44.07%) was selected because the dissolved Ca2+ could bind with SAP and restrain the initial swelling effect of SAP during mortar mixing [27,28]. The chemical compositions of cement and fly ash are given in Table 1. The SAP used in this experiment has a rough surface that minimizes swelling when exposed to an ionized solution. Fig. 1 shows secondary electron image (SEI) visualizations of SAP particles in dry condition. Most of the SAP particles have irregular shapes, and the particle sizes range from approximately 10–500 lm. The properties of the SAP are presented in Table 2. 2.2. Mix proportions design Eight series of mixes were developed by the absolute volume. The mass of water was fixed to 237 kg/m3 which gives the control Table 1 Chemical composition of Portland cement and fly ash. Chemical composition (mass%)

Cement

Fly ash

Silicon dioxide (SiO2) Aluminum oxide (Al2O3) Ferric oxide (Fe2O3) Calcium oxide (CaO) Magnesium oxide (MgO) Sulfur trioxide (SO3) Alkalis (Na2O) Chlorine (Cl) Sum (SiO2 + Al2O3 + Fe2O3)

21.2 5.2 2.8 64.2 – 2.0 0.65 0.004 29.2

26.6 10.96 9.05 44.07 1.85 5.36 – – 46.61

Fig. 1. SEM image of SAP particles.

Table 2 Properties of SAP. Material

Apparent density (g/cm3)

Water absorption (g/g)

SAP

0.7

417

3

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975 Table 3 Mix proportions of mortar. Specimen

Cement (kg/m3)

Fly ash (kg/m3)

Water (kg/m3)

Sand (kg/m3)

SAP (mass% cement)

F00 F15 F25 F35 F45 F45S4 F45S6 F45S8

524.7 446.0 393.5 341.1 288.6 288.6 288.6 288.6

0 78.7 131.2 183.6 236.1 236.1 236.1 236.1

237 237 237 237 237 237 237 237

1458.8 1458.8 1458.8 1458.8 1458.8 1458.8 1458.8 1458.8

0 0 0 0 0 4% 6% 8%

mix to have w/c of 0.45 while the effective w/c ratios were reduced with the increase in SAP replacement ratio as no additional water was added during mixing. The mass of binder was fixed to 524.7 kg/m3. Cement was partially replaced with fly ash in mass proportion of 15%, 25%, 35% and 45%. Fly ash content of 45% was selected to mix with SAP because its advantage over rheological properties and ability to conquer the initial swelling of SAP. SAP replacement ratios were used in a portion of 4%, 6% and 8% by mass of cement because it was reported by Chindasiriphan and Yokota [10] that the higher replacement ratio has higher potential to enhance self-healing capability. The mass of the fine aggregate was constant at 1458.8 kg/m3 for all mixes. The details on mix proportions are listed in Table 3. 2.3. Mixing and specimen preparation Fig. 3. The specimen subjected to splitting load.

Cement, sand, fly ash and half of the SAP were premixed for 1 min, then the remaining SAP was gradually added and mixing continued for 2 more minutes to obtain an almost uniform distribution of materials. Dry mixing was necessary to avoid adhesion and workability reduction of SAP after its exposure to water. Once the solid particles were well-mixed, water was added, and mixing was continued for 3 more minutes. Eight series of cement mortar were cast in cylindrical molds to produce specimens with a diameter of 100 mm and a height of 200 mm. After demolding, the specimens were cured in tap water for 28 days. The specimens were divided into two sets: The first set was subjected to the compressive strength test, and the second set was used to study selfhealing behavior. In the second set, each specimen was cut into four discs (50
100/ mm). The top and the bottom discs were discarded. The remaining middle discs were jacketed with PVC to control the crack width and prevent extensive damage (Fig. 2). Precracking was achieved by a universal testing machine (UTM). Two steel pins were engaged on each side to hold the specimen longitudinally on the UTM. A concentrated load was gradually applied through a steel plate placed on top of the specimen

(Fig. 3). During pre-cracking, the splitting load simultaneously deformed the PVC jacket and the specimen. The lateral confinement at failure generated pre-cracking that satisfied the control crack width of 0.2 ± 0.1 mm. The initial crack width was obtained by averaging the crack widths measured at 6 locations: 3 on the top surface and 3 on the bottom surface. The measuring points were 25 mm apart and 25 mm from the edge of the specimen. 2.4. Exposure conditions Two exposure conditions were used. Wet-dry exposure was used to simulate concrete cracking in a splash and tidal zone. Continuous water immersion was used to simulate healing in a submerged zone. Pre-cracked specimens were healed for 28 days under the conditions denoted below.  FXX (specimen with fly ash): immersed in water at 40 °C.  F45SX (specimen with fly ash and SAP): immersed in water at 40 °C.  F45SX_WD (specimen with fly ash and SAP): immersed in water at 40 °C for 1 day and air-cured at room temperature for 1 day. The procedure was repeated as a loop. The temperature of 40 °C was selected because it was reported [29] that fly ash develops a noticeable degree of cement hydration at early age, and the given temperature has no detrimental impact on long-term strength development. 2.5. Test procedure

Fig. 2. The specimen (pink) was mounted in a PVC pipe (gray) with epoxy. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2.5.1. Mechanical properties investigation The compressive strength was determined in accordance with ASTM C39 [30] to investigate the influence of the fly ash and SAP on the mechanical behavior of the cement mortar. The recovery of the crack’s water tightness was analyzed using a test of water discharge through the crack. The test was carried out periodically

4

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

during the post-curing stage using the apparatus shown in Fig. 4. It was specifically designed to apply a constant hydrostatic pressure sufficient to achieve water infiltration throughout the crack. The apparatus components include an inflow outlet that supplies tap water to the top of the apparatus, an overflow outlet that is designed to drain excessive inflow and to maintain the water head at 280 mm above the top surface of the specimen, and a measurement outlet at the bottom for measuring the mass of water that flowed out through the crack. At the end of the water discharge test, the width of the healed crack was obtained based on the method of the initial crack width measurement (see Section 2.3). The measurements were taken at the original marked points. Roig-Flores et al. [31] proposed a self-healing assessment of crack closure that uses the crack closing ratio denoted in Eq. (1).

Crack closing ratio ¼ 1 

Final crack width w28 ¥0 ¼1 Initial crack width w0

ð1Þ

2.5.2. Chemical and microstructural analyses An optical stereomicroscope was used to confirm the results of crack closure and identify the morphology of the newly formed self-healing materials along the healed crack. Image processing was used to enhance the contrast of the photo. The effect of fly ash and SAP inclusion was further analyzed by scanning electron microscopy (SEM). The samples were collected from the crack interface and inner cracking area of healed specimens. Each sample was trimmed to a 10 mm
10 mm
10 mm cube, oven-dried at 50 °C for 24 h, and then stored in a desiccator to avoid carbonation. Once the test began, the samples were mounted onto the SEM sample stage and impregnated with gold. An accelerating voltage of 10–20 kV was set as the operation condition. Thermogravimetric/differential thermal analysis (TG/DTA) was used to analyze newly formed self-healing materials found along the crack. The cement mortar around the crack was crushed to prepare cement powder. The particle size of the pulverized powder was reduced by using a planetary ball mill operated at 450 rpm for 10 min until a particle size of no more than 100 lm was obtained. The powder sample was heated at a constant heating rate of 10 °C/min until reaching 1000 °C. The mass loss due to the decomposition of materials that occurred with increasing temperature was recorded. The amount of portlandite (Ca(OH)2) was determined by the mass change at approximately 450–550 °C, as portlandite transforms into CaO and H2O [32,33], while the amount of calcium carbonate or calcite (CaCO3) was obtained through mass change during decarbonation that occurs around 650–900 °C [33,34]. Based on the molecular mass of portlandite (MWCH), the molecular mass of water (MWH20), the mass of the reference sample

(Wref) and the mass measured at two inflection points on the TG curve ðDW), the amounts of compounds were obtained as shown in Eq. (2), and a similar equation was used as Eq. (3) to determine the amount of calcite.

CHð%Þ ¼ DW


MW CH
W ref MW H2 O

CaCO3 ð%Þ ¼ DW


ð2Þ

MW CaCO3
W ref MW CO2

ð3Þ

3. Results and discussion 3.1. Mortar flow The mortar flow was measured according to ASTM C1437, a standard test method for the flow of hydraulic cement [35]. The test results are presented in Table 4. The flow diameter was found to increase with increase in fly ash replacement ratio. This agrees with a report from the American Coal Ash Association [36] stating that the smooth, spherical surface of fly ash particles improves the workability of fresh concrete, as the particles confer a lubricating effect by behaving like tiny ball bearings. For every 10% increase in fly ash content, the flow diameter was found to increase by approximately 1.9%. Conversely, the slump was found to decrease with increase in SAP content, and that decrease was notable at high SAP contents. F45S4 and F45S6 had 17% and 27% less workability than F45 had, and F45S8 had the lowest workability. The results show that sudden exposure to water transformed the SAP into an insoluble gel from which an adhesive effect was observed. Simultaneously, the SAP was responsible for absorbing free water, which produced a lower w/c ratio for the concrete system. 3.2. Compressive strength The results of compressive strength at 28 days of curing are summarized in Table 5. The compressive strength in each series was averaged with 3 specimens. According to the experiment results, the specimens with fly ash and SAP showed a tendency for compressive strength to be lower at SAP contents of greater than 4%. The averaged compressive strength of F45S4 showed approximately 19% decrease compared to that of F45, and those of F45S6 and F45S8 yielded about 28% and 25% decrease, respectively. The compressive strength reduction was validated by Lee et al. [17] and Hasholt et al. [37], who concluded that the initial swelling of SAP generates a reasonable amount of SAP voids which can be classified as macropores (>50 nm in diameter). SAP voids were generated through the initial swelling of SAP. Snoeck et al. [24] further explained that in fresh mix, macropores spontaneously form and become occupied by swollen SAP particles. Later, the concrete pore solution is consumed by cement hydration, which

Table 4 Flow diameters.

Fig. 4. The water discharge measurement apparatus.

Specimen

Flow diameter (mm)

Flow diameter difference compared to F00 (%)

F00 F15 F25 F35 F45 F45S4 F45S6 F45S8

194 195 195 199 202 167 148 101.6

0 0.5 0.5 2.6 4.0 13.9 –23.7 *

(*The specimen has no slump.)

5

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975 Table 5 Compressive strength at 28 days of curing. Type

F00 F15 F25 F35 F45 F45S4 F45S6 F45S8

Compressive Strength Strength (MPa)

Standard deviation (MPa)

Coefficient of variation (%)

Strength reduction (%)

55.4 54.2 54.8 52.2 49.1 40.0 35.3 36.7

0.35 0.64 1.52 1.34 1.23 2.20 2.62 0.87

0.63 1.19 2.78 2.57 2.50 5.50 7.42 2.38

0 2.16 1.03 5.79 21.28 27.70 36.31 33.79

reduces the ambient moisture at locations of SAP deposition. Subsequently SAP releases the entrapped water, causing the SAP to shrink. After SAP voids form, they result in increases in the total porosity of the concrete system. In contrast, some studies reported that concrete in presence of SAP possibly shows decrease or increase in strength depending on which side of the mechanism is dominant [10,37]. Strength loss predominantly results from porosity increases. Strength gain is characterized by a combination of the three mechanisms. First, strength increases with decrease in w/c ratio. Second, strength improvement is a function of the densification of the pore structure and the enhanced hydration degree predominated by an effective internal curing of SAP. Third, SAP mitigates the development of microcracks from self-desiccation. In cases of SAP content higher than 4%, it was found that the effect of the porous matrix is dominant, which results in compressive strength reductions at higher SAP contents. The measurement was carried out at 28 days of curing. The results showed that the partial substitution of cement with fly ash caused a slight reduction in compressive strength. F15, F25 and F35 showed a slight decrease in strength compared to F00. The strength decreased steeply with increase in the fly ash proportion as F45 resulted in approximately 20% of strength reduction compared to that of F00. However, the long-term strength was essentially greater than that of the control specimen, as the pozzolanic reaction in fly ash contributed to gradual strength development [23]. Therefore, the fully hydrated specimen containing fly ash is expected to attain a higher compressive strength than the control specimen. 3.3. Effect of fly ash on SAP replacement ratio As mention in section 2.1, the current study was conducted using fly ash containing reasonable amount of CaO (44.07%), which can be classified as class C fly ash according to ASTM C618 [38]. Fly ash C possesses pozzolanic properties and some cementitious properties. The benefit of significantly high calcium content in cement pore solution was well discussed by Moon et al. [39],

Huber [40], Lee et al. [28]], and Mechtcherine et al. [27] The presence of calcium ions in the cement pore solution increases the likelihood of Ca2+ binding with the carboxylic groups in acrylate chains of SAP—binding that subsequently reduces the initial osmotic pressure and restrains the absorbency of the SAP particles. The results showed that at 45% fly ash content, specimens could achieve the maximum allowable SAP replacement ratio up to 8%. This maximum replacement ratio is about 2 times higher than that for mortar with 25% fly ash content [10]. The replacement ratio of higher than 8% developed critical rheological properties and was not feasible to be casted because SAP absorbed all water presented in the mix. Initial swelling of SAP is expected to have disadvantages on rheological improvement and compressive strength development because it generates an adhesive effect and the development of SAP voids produced from swollen SAP gel. It is worth to mention that addition of a certain type of fly ash has the potential to suppress the initial swelling of SAP, thereby allowing the mix to achieve a higher SAP content which spontaneously increases a probability for SAP to contact with an opening crack. 3.4. Effect of admixture content on crack closure performance Crack widths of pre-cracked and healed specimens are given in Table 6. The method shown in Section 2.3 enabled the specimens to achieve the controlled crack widths of 0.18 to 0.32 mm. The crack closure ratio is used to determine the crack sealing efficiency obtained from Eq. (1). The experimental results associated with cracking load indicate that a slightly lower cracking load is required to open the crack at higher fly ash proportions. During healing, the crack was filled with newly formed white carbonatelike particles which primarily emerged from the crack surface boundary. The crack opening and crack closure at 28 days of healing are demonstrated in Fig. 5. Stereomicroscopic investigation (Fig. 6) shows the newly deposited material to be crystallized calcium carbonate. This crack-closure mechanism is supported by Ramm and Biscoping [41], Yang et al. [42] and Stuckrath et al.

Table 6 Cracking load and crack closure ratio at 28 days of curing. Specimen

Initial crack width (mm)

Final crack width (mm)

Cracking load (kN)

Crack closing ratio

F00 F15 F25 F35 F45 F45S4 F45S4_WD F45S6 F45S6_WD F45S8 F45S8_WD

0.32 0.26 0.22 0.20 0.21 0.20 0.18 0.19 0.19 0.18 0.19

0.18 0.07 0.03 0.04 0.01 0.03 0.06 0.05 0.05 0.00 0.00

50.6 50.9 47.1 48.8 49.3

0.45 0.73 0.86 0.80 0.95 0.84 0.67 0.76 0.75 1.00 0.98

37.1 40.8 40.3

6

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

Fig. 5. Examples of crack opening (left) and crack closure at 28 days of healing (right).

carbon dioxide exist. The dissolved calcium ions then bind to the carbon dioxide, resulting in the following crack-healing reactions.

Fig. 6. Stereomicroscope image of self-healing products deposited along an open crack.

[43], who proposed that the continuing hydration of unreacted cementitious material, the swelling of C-S-H, and the crystallization of calcium carbonate including the deposition of solid particles in water are the crucial phenomena that facilitate self-healing. It is observed that calcium carbonate is prominently consolidated at the open crack surface rather than at the inner section of the crack. Edvardsen [44] and Wu et al. [16], suggested that calcium hydroxide produced by cement hydration leaches out and substantially concentrates at the cracked surface where greater amounts of

H2 O þ CO2 $ H2 CO3 $ Hþ $ HCO3 $ 2Hþ þ CO2 3

ð4Þ

Ca2þ þ CO2 3 $ CaCO3 ðpHwater > 8Þ

ð5Þ

Ca2þ þ HCO3 $ CaCO3 þ Hþ ð7:5 < pHwater < 8Þ

ð6Þ

Fig. 7 demonstrates SAP configuration before soaking. The stereomicroscopic images were used to examine the crack-sealing mechanism attributed to the expansion of SAP. The mechanism was activated by the presence of water at the damage areas, where SAP absorbed water and swelled to seal the crack. Fig. 8(a) demonstrates that the crack is completely sealed by the expansion of SAP, whereas the SAP particles deposited at the location shown in Fig. 8 (b) show that crack widening is partially mitigated by the swollen gel, but that the SAP is unable to completely seal the open cracks. It is observed that specimens subjected to wet-and-dry cycles (denoted as ‘‘WD”) had lower average values of crack closing ratio. The results attest to the significance of water in promoting the selfhealing mechanism. In addition, the initial crack width greatly influences the crack-healing rate (the rate that determines the speed of the rehabilitation process) and the healing efficiency, as smaller cracks have higher potential to close completely regardless of the healing conditions. A tighter crack is characterized by concentrated Ca2+ deposition at the surface, which enhances the possibility of carbonation while requiring a lower amount of newly formed healing agent to bridge the interconnecting surface. However, the crack closing ratios are rather wide ranging, presumably

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

Fig. 7. Stereomicroscope image of dry SAP.

because the specimens had different degrees of initial damage. However, specimens with higher SAP content showed narrower standard deviations for crack closure ratios than those shown by specimens with less SAP or no SAP. It is noted that increasing the SAP content moderately improves the crack closure performance through the comprehensive internal curing that results from the release of entrapped water by SAP. 3.5. Effect of fly ash and SAP on water discharge through the crack Water discharge through the crack is plotted in Figs. 9–12. Crack closure resulting from the new formation of self-healing materials is responsible for reductions in such water discharge. The SAP-containing specimens show a significant reduction in water discharge after exposure to water. The average initial water discharge for the SAP-containing series was approximately 55% lower than that of F45. Among the SAP-containing specimens, F45S8 and F45S8_WD showed the initial water discharge of approximate of 72.1 mL/min, which is about one-quarter of F45. The results draw attention to the benefit of SAP and the significance of increasing the SAP dosage to promote instantaneous crack sealing after exposure to water. The sharp decline in water discharge resulted from the swelling of the SAP. After swelling, SAP acts as an internal water barrier, sealing the open channel. The results are consistent with the work of Lee et al. [17] and Snoeck et al. [45].

(a) Completely sealed crack by SAP

7

In the case of continuous water immersion, the specimens showed a gradual decrease in the water discharge rate after the first measurement. At the second and subsequent measurements, the SAP is expected to reach its maximum swelling capacity. Water discharge reduction beyond this measurement was characterized by hydrated products from pozzolanic reaction and further hydration of cementitious materials. In the series exposed to the wet-dry condition (denoted as WD), the specimens showed an inconsistent trend of water discharge through the crack. The trend in water discharge reduction was influenced by decreases in SAP osmotic pressure caused by inadequate water supply during air curing. In contrast, the change in the rate of flow through a crack for specimens in the WD series stabilized at approximately 12 days of healing. The authors believe that the water discharge rate at that period was regulated by the crack healing mechanism, whereby permanent healing products formed to seal the crack. The non-SAP specimens had much higher initial water discharge and less water discharge reduction. At the end of the observation, none of the specimens in this series had completely restored surface watertightness. Water discharge mitigation increased with increase in fly ash content, because the precracked specimens had narrower initial cracks. The addition of fly ash had no obvious effects on water discharge reduction. In contrast, F00 (the control specimen) had a rapid self-healing rate for the first 4 days before the rate fell to become similar to those of the other specimens. F00 later attained a final water discharge of about 250 mL/min. Because it has the highest cement content, F00 has a higher probability of its non-hydrated cement particles reacting with water during the first 4 days of the healing period. The rapid self-healing rate is supported by Neville [46,47], who claimed that the further hydration of unreacted cement particles is capable only for concrete at an early age. Therefore, the overall results indicate that permeability recovery and crack healing in the non-SAP series could be limited to the autogenous crack healing explained in Section 3.4. At 28 days of healing, F45 reached a final water discharge of roughly 50 mL/min; the lowest water discharge in the non-SAPcontaining series. In contrast, in the SAP-containing series, F45S8 developed a completely watertight crack after 12 days of healing. It is worth noting that the higher SAP content accounts for the enhanced water discharge reduction and the recovery of the specimen’s watertightness despite the exposure condition. The effect of high fly ash content on water discharge reduction is unclear. In contrast, specimens with higher fly ash content showed a smaller crack width under the same failure mode. Therefore, the influence

(b) Partially sealed crack by SAP

Fig. 8. Stereomicroscopy showing the crack sealing mechanism from SAP swelling.

8

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

800

600 500

Flow rate (mL/min)

700 Flow rate (mL/min)

250

F00 F15 F25 F35 F45

400 300 200

F45S8

200

F45S8_WD

150 100 50

100 0

0

0

4

8

12 16 Days elapsed

20

24

28

0

4

8

12 16 20 Days elapsed

24

28

Fig. 12. Water discharge through cracks for F45S8 / F45S8_WD.

Fig. 9. Water discharge through cracks for F00-F45.

250 Flow rate (mL/min)

F45S4 200

F45S4_WD

150 100 50 0 0

4

8

12 16 20 Days elapsed

24

28 Fig. 13. Correlation between initial crack width and initial discharge rate.

Fig. 10. Water discharge through cracks for F45S4/F45S4_WD.

250

F45S6

Flow rate (mL/min)

200

F45S6_WD

150 100 50

the pre-cracked specimens were prepared by a consistent method, the cracking patterns had many forms, and these forms were regulated by many factors, such as the aggregate arrangement inside the mortar structure and the failure mode of the specimens. The mentioned parameters influenced individual specimens to form unique crack geometries, patterns and crack surface roughnesses. Water is assumed to be discharged as incompressible laminar flow through a porous channel. In consideration of a permeability coefficient governed by Darcy’s law, the relation in Eq. (7) is found [45,48].

k¼

0 0

4

8

12 16 20 Days elapsed

24

28

Fig. 11. Water discharge through cracks for F45S6 / F45S6_WD.

of fly ash as a damage-prevention agent and its benefits regarding self-healing performance are profound.

3.6. Relationship between flow rate and crack width The correlations between initial crack widths and initial water discharge rates are shown in Fig. 13. It was found that even though two specimens had the same crack width, they could have different initial water-discharge rates. It was assumed that the liquid was discharged through a heterogeneous concrete matrix. Even though

  aL h0  ln A  tf hf

ð7Þ

where a is the cross-sectional area of the testing tower (m2), L is the thickness of the specimen (m), A is the specimen cross-sectional area (m2), t f is the monitoring time (s), h0 is the initial head (m) and hf is the measuring head (m). According to Tsukamoto and Woener [49], the relationship between permeability coefficient k and crack width w (m) can be obtained by the Eq. (8).

k¼

ag  ls  g h  w3  12v d

ð8Þ

where ag is a discharge rate coefficient specifying the crack surface roughness, ls is the length of the crack at a right angle to the flow direction (m), g is the gravitational constant (m/s2), v is the kinematic viscosity (m2/s), w is the crack width on the inlet side (m), h is the height of the testing tower (m) and d is the crack depth (m). Eq. (8) suggests that the crack width on the inlet side has the greatest influence on the permeability of the cracked specimens.

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

In the cases of the SAP-containing specimens, it was confirmed that the average initial water discharge rates were lower than those of the non-SAP containing specimens of comparable crack width. A maximum initial water flow rate of less than 200 mL/min was recorded for the SAP-containing series, whereas the maximum initial flow rate for the non-SAP specimens was roughly 1400 mL/min. It is noted that SAP absorbs water, increases in volume and is transformed into a watertight membrane. The relationship between healed crack width and final discharge rate is shown in Fig. 14. The figure shows a linear relationship between crack width and water discharge through the crack. The healed specimens have reduced crack cross-sectional areas, which results in decreased water discharge through the crack. The evidence suggests that assessments of self-healing should address both water discharge through the crack and crack width. Self-healing performance is found to significantly affect crack closure when the initial crack width is less than 0.25 mm. Nonetheless, the healing effect can be observed for cracks wider than 0.25 mm and the self-healing mechanism remained active at the end of the observation period. However, concrete becomes more vulnerable when a crack starts to propagate, because the crack serves as a pathway that increases the likelihood of harmful agents invading the concrete and exacerbating steel corrosion. Angst et al. [50] and Wang and Lu [51] discussed various parameters that influence the corrosion rate of reinforced concrete, including a wide range of threshold chloride concentrations and several environmental factors. Chlorine penetration is considered to be

Fig. 14. Correlation between healed crack width and final discharge rate.

9

an irreversible process, which may comprehensively deteriorate concrete structures. A crack must be sealed as soon as possible, which means the time required for healing should be minimized. In parallel with self-healing efficiency, it is also very interesting to focus more on the healing rate, and further research is required for a better understanding of durability. 3.7. Thermogravimetric analysis of self-healing performance The results of derivative thermogravimetric (DTG) curves are shown in Fig. 15. Self-healing is evaluated through an increase in calcite content and newly formed C-S-H. Four distinct peaks were generated. The first noticeable peak at approximately 60 °C, represents the decomposition of ettringite. The second, at 120 °C, represents that of monosulfoaluminate, followed by portlandite at 480 °C and calcite at approximately 680 °C. These peaks are in good agreement with Snoeck et al. [24]. The calculated amounts of portlandite and calcite are displayed in Fig. 16. F00 has a calcite content of 2.57%, versus 2.27% for F15, 1.91% for F25, 1.96% for F35 and 1.55% for F45, while F45S4 has a calcite content of 4.5%, versus, 5.8% for F45S4_WD, 5% for F45S6, 6.1% for F45S6_WD, 5.4% for F45S8 and 5.8% for F45S8_WD. It is noticeable that the specimens with SAP have roughly three times the calcite formation of the specimens without SAP. At higher SAP replacement ratios, slightly greater calcite formation is found. In addition, the results show that calcite content in the specimens healed under wet-dry cycles averages 1% higher than that in the same type of specimens healed under continuous water immersion. A statistical analysis by Hills et al. [52] proposed that there are a few environmental variables influencing carbonation rate. For example, the completely liquid filled pores in concrete reduce the diffusion rate because gases are transported in air 104 times as rapidly as in water. However, some humidity is essential for the dissolution of calcium ions and carbon dioxide, as discussed in Eqs. (4), (5) and (6). A suitable condition for carbonation occurs at relative humidity of 50–70% [52,53]. The portlandite content was the highest in F00 (1.52%) and the lowest in F45 (0.49%). In the presence of fly ash, specimens undergo a reduction in portlandite proportion. All the SAP-containing specimens achieved a portlandite content of 1% regardless of the SAP replacement. However, the reason why the portlandite content of the SAP-containing specimens diverged from that of F45 remains unclear, but it might be expected by presence of fly ash and SAP. Further study on this

Fig. 15. DTG curves for ettringite, portlandite and calcite. (a) Specimens with fly ash. (b) Specimens with fly ash and SAP.

10

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

F45 0.49

1.55

F35 0.57 F25 0.70 F15

0.82

F00

1.96 1.91 2.27

1.52

0.00

2.57 2.00

4.00

Portlandite (%w)

6.00

8.00

Calcite (%w)

F45S8_WD

0.9

5.8

F45S8

1.0

5.4

F45S6_WD

1.0

F45S6

1.0

F45S4_WD

1.2

F45S4

6.1 5.0 5.8

0.8 0.0

4.5 2.0 Portlandite (%w)

4.0

6.0

8.0

Calcite (%w)

Fig. 16. Thermogravimetric analysis results [mass%] for portlandite content and calcite content.

Fig. 17. SEM image of the healed crack surface.

Fig. 19. SEM image of ITZ.

Fig. 18. EDS analysis results for the healed crack surface. Fig. 20. SEM image of SAP void.

reason is required. It is worth noting that the reduction in portlandite content is characterized by a pozzolanic reaction in which fly ash compounds react with Ca(OH)2 in concrete to produce C-SH that subsequently contributes to the gradual development of compressive strength, the densification of the pore structure and self-healing. 3.8. Microstructural analysis by SEM Microstructure images were captured by SEM with an EDS analyzer. This analyzer is used to investigate the morphology of newly formed healing materials and to determine their chemical compositions. The observation confirmed that the main components of the hydrated products were calcium and silicon compounds. It is worth noting that the forms of the self-healing materials differed

with respect to the location of the self-healing agent, the concrete depth, and the confinement and exposure conditions. Fig. 17 shows an image of the external crack surface of F45S8. The newly hydrated products have the morphology of needles growing on rhombohedral crystals. The needle crystals have estimated lengths of approximately 20–50 mm and estimated diameters of 200–500 nm. They are observed growing in multiple directions over a layer of what appear to be calcite. From their morphology, the needles can be identified as aragonite (CaCO3 has three morphologies: aragonite, vaterite and calcite). The quantitative data of the EDS analysis is presented in Fig. 18. Multiple layers of self-healing materials have formed on the crack surface,

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

which has resulted in refraction and absorption of the X-ray beams. From the EDS analysis, we were unable to precisely identify the chemical composition of the self-healing products. Fig. 19 demonstrates that the densification of the cement matrix is visible at the interfacial transition zone of F45S4. Fig. 20. demonstrates the location of an SAP void on the crack surface of F45S8_WD.The small yellow rectangle indicates a crystal particle of approximate size of 80 lm that appears to be SAP. The efficacy of SAP as a water retaining agent, its ability to supply water to unhydrated materials and its benefits in self-healing are proven.

11

Fly ash was observed at the inner cracked boundary of F45S8. In Fig. 21, new healing products are observed developing from the surface of a fly ash particle, where they are growing to fill the crack cavity. The EDS results indicated that the primary hydrated products consist of high quantities of oxygen (O), carbon (C), calcium (Ca) and silicon (Si), and lower quantities of sodium (Na), aluminum (Al), magnesium (Mg) and iron (Fe) (Fig. 22). In consideration of the Si and Ca contents, Fig. 23 (a) shows that C-S-H is found concentrated around the fly ash particles , while calcium carbonate is present in significant amounts around the outer region (Fig. 23 (b)). This is in line with the results of Siad et al. [54], who further suggested that the Ca/Si ratio can be used to predict C-S-H stiffness. Self-healing products tend to be stiffer at lower Ca/Si ratios. It is worth mentioning that in the presence of fly ash, denser and more durable self-healing products are found. In Fig. 21, it is clearly visible that at 28 days of healing, most of the fly ash particles have not completely reacted. This suggests the possibility of self-healing activities under pozzolanic reaction in the later age. The assumption is supported by the study of Hung and Su [55], who proved the reactivity of fly ash by comparing the silicon peak of a fly ash specimen healed underwater for 28 days to that of the same specimen healed underwater for 90 days. Their silicon peak is higher at 90 days of healing, which suggests that the amount of new C-S-H increases with healing time. 4. Conclusions

Count per second (CPS)

Fig. 21. SEM image of fly ash around the healed inner crack boundary.

Fig. 22. EDS analysis results around the healed inner crack boundary.

(a) C-S-H in green

This study investigated on how fly ash and SAP promote the self-healing ability of concrete. It discussed the effects of both materials on rheological behavior, compressive strength, crack closure performance, carbonation and the recovery of concrete permeability. The experimental program focused on the effect of fly ash and SAP on the water discharge obtained at 28 days of healing in two different exposure conditions: with wet-dry cycles and with continuous water immersion. From the experimental results, the following conclusions can be drawn: 1. SAP is responsible for slump reduction, which results from the adhesive effect of swollen SAP and the absorption of free water in the concrete system. The initial absorbency of SAP can be suppressed by adding concentrated Ca2+ to the concrete mixing solution, as it binds to the carboxylic groups in acrylate chains of SAP which leads to the desorption of stored liquid. 2. The addition of fly ash improves damage prevention. Specimens containing fly ash obtain finer cracks under cracking loads than specimens without fly ash. In addition, the selection of fly ash with a high calcium content has the potential to enable mix proportions with higher SAP replacement

(b) CaCO3 in red Fig. 23. SEM image around a fly ash particle.

(c) C-S-H – CaCO3

12

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975

3.

4.

5.

6.

7.

8.

9.

10.

ratios. It is also worth noting that the pozzolanic reaction in fly ash has a great potential to enhance the self-healing performance that results from the development of newly formed C-S-H products. The compressive strength tends to decrease with increases in SAP dosage beyond 4 mass% of cement. The strength development at 28 days is interfered with by fly ash. The compressive strength of F45 is approximately 20% lower than cement mortar without fly ash. Specimens that were healed in water immersion achieve greater self-healing than those subjected to wet-dry cycles, because the presence of water enhances the hydration of fly ash and unhydrated cementitious materials, which subsequently promotes self-healing. The initial water discharge is approximately 55% lower for the SAP-containing specimens than for the non-SAP specimens. The crack sealing mechanism for the SAP-containing specimens is promoted by the formation of an impermeable gel that expands to seal an opening crack which occurs instantaneously after the SAP is exposed to water. Increasing the SAP replacement ratio improves the crack sealing performance, as characterized by greater crackwidening mitigation and water discharge reduction resulting from extensive moisture supply from the SAP. Meanwhile, the SAP greatly accelerates carbonation activities which enhances self-healing performance by introducing a higher degree of calcium carbonate deposition along the open crack compared to non-SAP containing specimens, which agrees well with the results of water discharge through the crack. The initial crack width has a significant influence on the selfhealing rate and self-healing threshold. At 28 days of selfhealing, the specimens with initial crack widths under 0.25 mm achieve satisfactory crack closure. The healed specimens develop permanent hydrated substance at the damage area. The SEM-EDS analysis results of healed specimens show that calcium carbonate is found in significant amounts around the surface and outer region of the healed specimens while the C-S-H highly concentrates around fly ash particles at the inner depth. Midterm self-healing ability of concrete can be enhanced by the coupling effect of fly ash and SAP. The crack closure is found to be associated with the development of selfhealing products such as calcium carbonate and C-S-H. To tackle budgetary constraints and reduce life cycle cost of concrete structures, it would be interesting to further evaluate on cost-benefit analysis of self-healing concrete made with fly ash and SAP in the future which should be carried out along with the investigation on structural aspects, and the long-term efficiency of SAP.

Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors would like to thank Nippon Shokubai Co., Ltd. for supplying the SAP. We wish to express our gratitude to Professor Minoru Kunieda of Gifu University and his team for all their support. A portion of this work was conducted at Thin Section Laboratory, Faculty of Science of Hokkaido University and the XPS

Analysis Laboratory, a joint-use facility of Hokkaido University, supported by the Material Analysis and Structure Analysis Open Unit (MASAOU).

References [1] A. Tom, Self-healing concrete could save £40bn in maintenance costs, (2017). http://www.materialsforengineering.co.uk/engineering-materials-news/selfhealing-concrete-could-save-40bn-in-maintenance-costs/109230/ (accessed October 8, 2018). [2] D. Gardner, R. Lark, T. Jefferson, R. Davies, A survey on problems encountered in current concrete construction and the potential benefits of self-healing cementitious materials, Case Stud. Constr. Mater. 8 (2018) 238–247. [3] H.X.D. Lee, H.S. Wong, N. Buenfeld, Self-sealing cement-based materials using superabsorbent polymers, in: Inter. RILEM Conf. on Use of Superabsorbent Polymers and Other New Additives in Concrete, Copenhagen, Denmark, 2010, pp. 171-178. [4] N. Hearn, Self-sealing, autogenous healing and continued hydration: What is the difference?, Mater. Struct. 31 (1998) 563–567. [5] H.W. Reinhardt, M. Jooss, Permeability and self-healing of cracked concrete as a function of temperature and crack width, Cem. Concr. Res. 33 (2003) 981– 985. [6] H. Liu, Q. Zhang, C. Gu, H. Su, V. Li, Self-healing of microcracks in engineered cementitious composites under sulfate and chloride environment, Constr. Build. Mater. 153 (2017) 948–956. [7] P. Escoffres, C. Desmettre, J.P. Charron, Effect of a crystalline admixture on the self-healing capability of high-performance fiber reinforced concretes in service conditions, Constr. Build. Mater. 173 (2018) 763–774. [8] S. Igarashi, A. Hosoda, T. Hitomi, K. Imamoto, Technical Committee on Repairing Technology in Cement-based Materials, Japan Concrete Institute (2011) 1–24. [9] Y. Yang, M.D. Lepech, E. Yang, V.C. Li, Autogenous healing of engineered cementitious composites under wet–dry cycles, Cem. Concr. Res. 39 (2009) 382–390. [10] P. Chindasiriphan, H. Yokota, Self-healing ability of concrete made with fly ash and superabsorbent polymer, in: 2nd ACF Symp 2017 Innovation of Sustainable Concrete Structures, Chiang Mai, Thailand, 2017. [11] P. Termkhajornkit, T. Nawa, K. Kurumisawa, Effect of water curing conditions on the hydration degree and compressive strengths of fly ash – cement paste, Cem. Concr. Compos. 28 (2006) 781–789. [12] K. Van Tittelboom, N. De Belie, F. Lehmann, C.U. Grosse, Acoustic emission analysis for the quantification of autonomous crack healing in concrete, Constr. Build. Mater. 28 (2012) 333–341. [13] K. Van Tittelboom, N. De Belie, D. Van Loo, P. Jacobs, Self-healing efficiency of cementitious materials containing tubular capsules filled with healing agent, Cem. Concr. Compos. 33 (2011) 497–505. [14] K. Van Tittelboom, N. De Belie, W. De Muynck, W. Verstraete, Use of bacteria to repair cracks in concrete, Cem. Concr. Res. 40 (2010) 157–166. [15] S.R. White, N.R. Sottos, P.H. Geubelle, J.S. Moore, M.R. Kessler, S.R. Sriram, E.N. Brown, S. Viswanathan, Autonomic healing of polymer composites, Nature. 409 (2001) 794–797. [16] M. Wu, B. Johannesson, M. Geiker, A review: Self-healing in cementitious materials and engineered cementitious composite as a self-healing material, Constr. Build. Mater. 28 (2012) 571–583. [17] H.X.D. Lee, H.S. Wong, N.R. Buenfeld, Self-sealing of cracks in concrete using superabsorbent polymers, Cem. Concr. Res. 79 (2016) 194–208. [18] O.M. Jensen, P.F. Hansen, Water-entrained cement-based materials: II. Experimental observations, Cem. Concr. Res. 32 (2002) 973–978. [19] S. Riyazi, J.T. Kevern, M. Mulheron, Super absorbent polymers (SAPs) as physical air entrainment in cement mortars, Constr. Build. Mater. 147 (2017) 669–676. [20] H.M. Jonkers, A. Thijssen, G. Muyzer, O. Copuroglu, E. Schlangen, Application of bacteria as self-healing agent for the development of sustainable concrete, Ecol. Eng. 36 (2010) 230–235. [21] M. Harilal, V.R. Rathish, B. Anandkumar, R.P. George, M.S.H.S. Mohammed, J. Philip, G. Amarendra, High performance green concrete (HPGC) with improved strength and chloride ion penetration resistance by synergistic action of fly ash, nanoparticles and corrosion inhibitor, Constr. Build. Mater. 198 (2019) 299–312. [22] J. Yu, C. Lu, C.K.Y. Leung, G. Li, Mechanical properties of green structural concrete with ultrahigh-volume fly ash, Constr. Build. Mater. 147 (2017) 510– 518. [23] M. Thomas, Optimizing the Use of Fly Ash in Concrete, in, Portl. Cem. Assoc. (2007). [24] D. Snoeck, L.F. Velasco, A. Mignon, S. Van Vlierberghe, P. Dubruel, P. Lodewyckx, N. De Belie, The effects of superabsorbent polymers on the microstructure of cementitious materials studied by means of sorption experiments, Cem. Concr. Res. 77 (2015) 26–35. [25] P. Termkhajornkit, T. Nawa, Y. Yamashiro, T. Saito, Self-healing ability of fly ash-cement systems, Cem. Concr. Compos. 31 (2009) 195–203. [26] Z. Zhang, S. Qian, H. Ma, Investigating mechanical properties and self-healing behavior of micro-cracked ECC with different volume of fly ash, Constr. Build. Mater. 52 (2014) 17–23.

P. Chindasiriphan et al. / Construction and Building Materials 233 (2020) 116975 [27] V. Mechtcherine, E. Secrieru, C. Schröfl, Effect of superabsorbent polymers (SAPs) on rheological properties of fresh cement-based mortars — Development of yield stress and plastic viscosity over time, Cem. Concr. Res. 67 (2015) 52–65. [28] H.X.D. Lee, H.S. Wong, N.R. Buenfeld, Effect of alkalinity and calcium concentration of pore solution on the swelling and ionic exchange of superabsorbent polymers in cement paste, Cem. Concr. Compos. 88 (2018) 150–164. [29] Y. Maltais, J. Marchand, Influence of curing temperature on cement hydration and mechanical strength development of fly ash mortars, Cem. Concr. Res. 27 (1997) 1009–1020. [30] ASTM C 39, Standard test method for compressive strength of cylindricalcompressive specimens, ASTM International, West Conshohocken, PA. [31] M. Roig-Flores, F. Pirritano, P. Serna, L. Ferrara, Effect of crystalline admixtures on the self-healing capability of early-age concrete studied by means of permeability and crack closing tests, Constr. Build. Mater. 114 (2016) 447– 457. [32] M. Henry, K. Hashimoto, I.S. Darma, T. Sugiyama, Cracking and chemical composition of cement paste subjected to heating and water re-curing, J. Adv. Concr. Technol. 14 (2016) 134–143. [33] L. Alarcon-Ruiz, G. Platret, E. Massieu, A. Ehrlacher, The use of thermal analysis in assessing the effect of temperature on a cement paste, Cem. Concr. Res. 35 (2005) 609–613. [34] E. Oniyama, P.G. Wahlbeck, Application of transpiration theory to TGA data : calcium carbonate and zinc chloride, Thermochim. Acta. 250 (1995) 41–53. [35] ASTM C 1437, Standard Test Method for Flow of Hydraulic Cement Mortar, ASTM International, West Conshohocken, PA. [36] American Coal Ash Association, Fly Ash Facts for Highway Engineers, (2003). https://www.fhwa.dot.gov/pavement/recycling/fach00.cfm (accessed October 9, 2018). [37] M.T. Hasholt, M.H.S. Jespersen, O.M. Jensen, Mechanical properties of concrete with SAP part I: development of compressive strength, in: Inter. RILEM Conf. on use of superabsorbent polymer and other new additives in concrete, Copenhagen, Denmark, 2010, pp. 117 - 126. [38] ASTM C618, Standard Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use in Concrete, ASTM International, West Conshohocken, PA. [39] J. Moon, S. Kang, S. Hong, Ionic dependence of behavior of water-entraining admixture in early age concrete, in: 2nd ACF Symp. 2017 Innovation of Sustainable Concrete Structures, Chiang Mai, Thailand. 2017.

13

[40] K. Huber, Calcium-induced shrinking of polyacrylate chains in aqueous solution, J. Phys. Chem. 97 (1993) 9825–9830. [41] W. Ramm, M. Biscoping, Autogenous healing and reinforcement corrosion of water-penetrated separation cracks in reinforced concrete, Nucl. Eng. Des. 179 (1998) 191–200. [42] Y. Yang, E. Yang, V.C. Li, Autogenous healing of engineered cementitious composites at early age, Cem. Concr. Res. 41 (2011) 176–183. [43] C. Stuckrath, R. Serpell, L.M. Valenzuela, M. Lopez, Quantification of chemical and biological calcium carbonate precipitation: performance of self-healing in reinforced mortar containing chemical admixtures, Cem. Concr. Compos. 50 (2014) 10–15. [44] C. Edvardsen, Water permeability and autogenous healing of cracks in concrete, ACI Mater. J. 96 (1999) 448–454. [45] D. Snoeck, K. Van Tittelboom, S. Steuperaert, P. Dubruel, N. De Belie, Selfhealing cementitious materials by the combination of microfibres and superabsorbent polymers, J. Intell. Mater. Syst. Struct. 25 (2014) 13–24. [46] A.M. Neville, Properties of concrete, 5 ed., Pearson, London, UK, 2011. [47] A.M. Neville, Autogenous healing - a concrete miracle, Concr. Int. (2002) 76– 82. [48] H. Song, S. Kwon, Permeability characteristics of carbonated concrete considering capillary pore structure, Cem. Concr. Res. 37 (2007) 909–915. [49] M. Tsukamoto, J.D. Woener, Permeability of cracked fibre-reinforced concrete, Darmstadt Concr. 6 (1991) 123–135. [50] U. Angst, B. Elsener, C.K. Larsen, Ø. Vennesland, Critical chloride content in reinforced concrete — a review, Cem. Concr. Res. 39 (2009) 1122–1138. [51] W. Wang, C. Lu, Time-varying law of rebar corrosion rate in fly ash concrete, J. Hazard. Mater. 360 (2018) 520–528. [52] T.P. Hills, F. Gordon, N.H. Florin, P.S. Fennell, Statistical analysis of the carbonation rate of concrete, Cem. Concr. Res. 72 (2015) 98–107. [53] S.K. Roy, D.O. Northwood, K.B. Poh, Effect of plastering on the carbonation of a 19-year-old reinforced concrete building, Constr. Build. Mater. 10 (1996) 267– 272. [54] H. Siad, M. Lachemi, M. Sahmaran, H.A. Mesbah, K.A. Hossain, Advanced engineered cementitious composites with combined self-sensing and selfhealing functionalities, Constr. Build. Mater. 176 (2018) 313–322. [55] C. Hung, Y. Su, Medium-term self-healing evaluation of engineered cementitious composites with varying amounts of fly ash and exposure durations, Constr. Build. Mater. 118 (2016) 194–203.