Abstract
Roller-compacted concrete pavement (RCCP) is considered superior to other pavement types with reference to cost, ease of construction, and performance. However, the aggregate demand is significantly higher in RCCP than in the conventional concrete pavement. It is predicted that natural minable limestone sources would be exhausted in India in the next 30 to 40 years. One way to reduce the natural aggregates (NA) requirements in RCCP is through the integration of recycled concrete aggregates (RCA). However, the physical property of RCA is significantly inferior to that of NA owing to the presence of adhered mortar (AM), which increases the water demand by 2.3 to 4.6 times and affects the compactness and hardened-state behavior of RCCP. This study has tried to enhance the compactness and improve the performance of RCCP containing RCA (coarse, fine, and total RCAs) through different synergetic approaches. This, in turn, enhances the interlocking capacity using the particle packing approach, followed by mitigating the negative effects of AM by altering the moisture states, then improving the compactability and lubricating the matrix with superplasticizers. The results depict that altering the moisture states alone could adversely affect the RCCP performance because of the moisture transfer mechanism from the hydraulic gradient. Moreover, the inclusion of superplasticizers in different moisture states could manifest better aggregate rearrangement and compactability. Further, it could improve the tensile behavior of the RCCP compared with the concrete containing NA. These findings favor the complete replacement of NA by RCA for low-volume rural road construction.
Keywords
Roller-compacted concrete pavements (RCCP) has numerous benefits, primarily construction cost owing to low binder content, high paving speed, formwork-less construction, and dowel-free joints ( 1 – 3 ). The strength development in RCCP occurs through the combined effects of aggregate-to-aggregate interlocking and binder–mortar stiffness ( 1 , 4 ). This is achieved by denser aggregate packing, resulting in fewer voids, and thus, less binder demand ( 5 ). Many particle packing approaches are available that could be followed to ensure denser aggregate interlocking in RCCP ( 6 ). Among all, the modified Andreassen, Fuller–Thomson, compressible packing, and solid suspension models produced RCCP mixtures with better performance ( 5 , 7–10).
Considering the higher aggregate demand for RCCP and a parallel shortage of the same, the use of recycled concrete aggregates (RCA) from construction and demolition waste seems to be a sensible way to induce sustainability in the highway sector ( 11 , 12 ). However, the presence of low-density porous adhered mortar (AM) around RCA could significantly affect the compactness, and thus, the aggregate interlocking capacity of RCCP; the compactness is usually determined by maximum dry density (MDD) and Vebe density. Based on the available literature, it is confirmed that the density of these hybrid mixtures could be lower than conventional RCCP in the range of 2.2% to 4.7% because of AM ( 7 , 13 , 14 ). Also, the water demand for these mixtures could be 8.3% to 35% higher than conventional mixtures for coarse RCA inclusion ( 7 , 13 , 14 ). This can be attributed to higher AM content in coarse RCA, which increases the water absorption characteristics ( 2 ). As a strong correlation exists between the compactness and strength characteristics of RCCP, the use of RCA could lead to a reduction in the compressive strength and indirect tensile strength by 5% to 35% and 18% to 23%, respectively, for the conventional RCCP ( 7 , 13 , 14 ). These studies strongly promote the treatment of RCA before using for RCCP; however, no such attempt has been made till date for such stiff mixtures. Nevertheless, the inferences from the studies conducted on conventional concrete could be adopted for enhancing the performance of RCA-inclusive RCCP, as done in the present study.
The RCA treatment techniques can be classified into three types: (i) changing the rheological parameters of the mixture through moisture adjustment, (ii) AM strengthening through cementitious/filler materials, and (iii) AM removal through physio-chemical techniques ( 2 , 15–17). The latter two techniques have been regarded as effective in enhancing the RCA quality but are energy-intensive ( 18 – 20 ), whereas the moisture adjustment technique is considered easy and can be applied to large-scale use. The moisture adjustment could be made either by fully/partially soaking the RCA or by adding mixing water at different intervals ( 21 – 25 ). Soaking the RCA beforehand could minimize the negative impact of AM on the fresh-state properties; however, it may or may not influence the hardened-state behavior of the concrete ( 25 – 29 ). For instance, previous studies ( 26 – 28 ) reported that both fresh and hardened properties were enhanced when RCA was pre-soaked up to 80% of its capacity compared with dry and fully soaked RCA states. In contrast, Poon et al. ( 25 ) observed better strength characteristics when air-dried RCA was used. Similarly, it was noted that fully dried aggregates could exhibit higher slump value, but the workability decreases drastically with time because of the absorption of water by the AM ( 25 ). On the other hand, altering the mixing sequence and duration could improve the rheological parameters and the interfacial transition zone (ITZ) of the RCA-inclusive concrete mixtures. For instance, two-stage mixing was found to enhance the RCA quality because of the filling of cracks and voids by cement paste during the mixing stage ( 30 , 31 ). However, the mixing duration was observed to be inadequate to achieve the equilibrium state of RCA absorption before the concrete specimens fabrication ( 32 ). This was owing to the use of RCA in the dry state, which releases air bubbles when the mixing water is absorbed, which could form macropores in the ITZ ( 32 ). These findings promote the adoption of a moisture-altering approach for RCA-inclusive mixtures; however, for stiff mixtures such as RCCP, the same has to be validated without affecting the aggregate interlocking mechanism.
Research Significance
From the existing knowledge available in the literature, it can be concluded that tailoring the rheological behavior of mixtures by amending the moisture state of RCA could be a prudent way to induce sustainability in RCCP. However, no such study is available till date due to the complex rheology of RCCP, wherein the compactness of the fresh mixes plays the most dominating role in affecting the structural integrity of the concrete. Similarly, lubrication of the RCCP through chemical admixtures is also an emerging area. Also, to date, no study has attempted to use the fine RCA for RCCP. The present study is a novel approach exploring the potential of RCA for RCCP by increasing the interlocking of aggregates, and thus, the compactness of the fresh mixtures through different concurrent approaches, namely, particle packing approach, altering the moisture states of the RCA, and lubricating and water reduction with the use of chemical admixtures. The main aim of this study is large-scale use of both the fractions of RCA (i.e., coarse and fine). Normally, fine RCA is not preferred because of its inferior quality, and this might significantly affect the compactness of RCCP. Also, this study is the first of its kind to simulate field compaction in the laboratory by using a dedicated concrete gyratory compactor for fabricating these hybrid mixtures.
Experimental Program
Materials
Three sizes of aggregates (both natural and RCA), in the range of 19 to 9.5 mm (0.74–0.37 in.), 9.5 to 4.75 mm (0.37–0.18 in.), and 4.75 to 0.075 mm (0.18–0.029 in.) were used for this investigation. The RCA was obtained from a nearby recycling plant in Chennai, India. The physical properties of the considered aggregates are shown in Table 1 ( 33 , 34 ). As seen in Table 1, all the RCA sizes were found to be inferior to natural aggregate (NA) sizes because of the higher AM content of 29.6% and 30% for 19 mm (0.74 in.) and 9.5 mm (0.37 in.), respectively. As a result of this, lower specific gravity and higher water absorption of 2.29 and 5.3%, 2.21 and 6.1%, and 2.18 and 8.8% were obtained for 19 mm (0.74 in.), 9.5 mm (0.37 in.), and fine RCA. Ordinary Portland cement (OPC) of grade 53 conforming to IS 12269 ( 35 ) was used throughout the study. Cement dosage of 300 kg/m3 (18.7 lb/ft 3 ) was kept constant for preparing all the RCCP mixtures. The chemical composition and physical properties of OPC are given in Tables 2 and 3, respectively ( 35 , 36 ).
Physical Properties of Aggregate
Note: na = not applicable.
Chemical Composition of Ordinary Portland Cement
Physical Characteristics of Ordinary Portland Cement
Optimization of Aggregate Fractions
In this study, the aggregate proportions of both RCA and NA were determined by experimental packing studies. A modified Vebe apparatus was selected to measure the packing density of aggregate mixtures to avoid the wall effect. A surcharge load of 22.7 kg (50 lb) was used to simulate the compaction effort of rollers as per ASTM C1170 ( 37 ). Initially, packing density was determined for NA and RCA at different replacement levels (coarse, fine, and total RCAs) for different combinations of aggregate mixtures (19 mm [0.74 in.], 9.5 mm [0.37 in.], and fine aggregate). Based on the individual packing density, ternary plots were developed to determine the optimal aggregate fractions that can produce a denser aggregate skeleton (discussed in the results and discussion section). The compaction process and packing density test procedure are as follows.
The required total mass to fill the Vebe mold was obtained from the ternary fractions of aggregate (19 mm [0.74 in.], 9.5 mm [0.37 in.], and fine aggregate).
The ternary mixtures were blended to obtain homogeneous mixtures.
The blended aggregates were poured into the Vebe apparatus and compacted for 45 s ( 38 ), and subsequently, the final height was noted as H.
By knowing the weight and specific gravity of individual aggregate size fractions and the volume of Vebe apparatus, the bulk density and packing density of ternary mixtures can be determined as follows:
where D is the diameter of Vebe mold; H is the height of compacted blended mixtures; P1, P2, and P3 are the coarse and fine aggregate fractions; SG1, SG2, and SG3 are the specific gravity of the corresponding aggregates.
The experimental packing densities for NA and coarse, fine, and total RCAs are illustrated in Figure 1. The variation in the packing density for the ternary mixtures is represented through the contour diagrams and color gradients (Figure 1). For example, the red color shows a higher magnitude of packing density, whereas the green and blue colors represent the medium and lower magnitude of packing density, respectively (Figure 1). As seen in Figure 1, whatever the replacement levels (partial and total replacement), the inclusion of RCA at their optimal aggregate fractions (at which the maximum packing density was achieved) rendered a similar packing density (0.81–0.82) to the NA blend (0.81). The selected optimal aggregate quantity for the considered RCCP mixtures is given in Table 4.

Ternary packing density diagram: (a) NA, (b) coarse RCA, (c) fine RCA, and (d) total RCA.
Mix Proportion of Control and RCA Mixtures
Note: na = not applicable; NA = natural aggregates; RCA = recycled concrete aggregates; OMC = optimum moisture content; w/c ratio = water–cement ratio; SP = superplasticizer; SSD = saturated surface dry. 1 kg/m3 = 0.062 lb/ft 3 ; and 1 MPa = 145.038 pounds per square inch.
Mixture Proportioning and Mixing Procedure
Particle packing approach is usually adopted for stiff mixtures to reduce the voids and increase the aggregate interlocking ( 10 ), and the same approach is followed for the present study. The aggregate quantity determined for all the RCCP mixtures based on the particle packing approach is given in Table 4.
For the present study, ACI 327 ( 39 ) and IRC: SP:68 ( 40 ) specifications, which are widely followed for RCCP, are considered for designing the mixtures (with respect to moisture content). These specifications suggest adopting the soil compaction methodology to determine the optimum moisture content (OMC) by developing the moisture density plot, as per ASTM D1557 ( 41 ). Initially, a control mixture was fabricated based on the optimum water content. Similarly, RCA–RCCP mixtures were produced at the dry state, saturated surface dry (SSD) state, dry state + superplasticizer (SP), and SSD state + superplasticizer (SP); in each mixture, the considered NA was replaced at 100% by a volumetric fraction of coarse, fine, and total RCAs. A polycarboxylate ether-based superplasticizer (SP) was used in superplasticizer (SP)-inclusive mixtures, and its dosage was selected based on the desirable Vebe time (20–40 s). The mix proportions for the considered RCCP mixtures are presented in Table 4.
In this study, the coarse and fine aggregates were added and mixed homogeneously for the first two min. in a pan mixer, followed by the inclusion of 20% mixing water into these dry materials. Afterward, total cement was added and mixed continuously for 1 min. Later, the remaining 80% water and superplasticizer were mixed and added to the mixture (cement + coarse aggregate + fine aggregate + 20% of mixing water) and continuously mixed for the next 3 min. Figure 2 schematically represents the mixing sequence and time duration in the present study.

Mixing procedure and time duration of roller-compacted concrete pavement (RCCP) mixtures.
Specimen Preparation and Test Methods
As soon as concrete mixtures were produced in a pan mixer, the Vebe consistency of RCCP mixtures was determined according to ASTM C1170 ( 37 ), where 13.5 kg (29.8 lb) of RCCP mixtures were placed and compacted until the formation of a mortar ring or a maximum of 60 s (whichever comes earlier). Subsequently, RCCP specimens were fabricated using the gyratory compaction techniques to mimic the actual field compaction energy through the vertical compaction pressure and kneading effort. The compaction parameters used in the present study conformed to ASTM C1800 ( 42 ), where the suggested internal angle, compaction pressure, and the number of gyrations are 1.16°, 600 kPa (87 pounds per square inch), and 60 (recommended range 50–70 gyrations), respectively. At 28 days, three cylindrical specimens were tested for compressive strength and indirect tensile strength, as per ASTM C39 ( 43 ) and ASTM C496 ( 44 ), respectively. Water absorption and porosity were evaluated from eight sliced specimens (100 mm [3.93 in.] [diameter] × 50 mm [1.97 in.] [height]), in accordance with ASTM C642 ( 45 ) and CPC 11.3 (RILEM) ( 46 ), respectively.
Results and Discussion
Packing Density
The optimum aggregate proportions and corresponding packing density for NA and RCA are shown in Figure 3. The X and Y axes of Figure 3 represent the natural & RCA replacement type and aggregate proportions, respectively of considered coarse and fine aggregates to achieve the maximum compactness. The maximum packing densities of NA, coarse RCA, fine RCA, and total RCA were found to be 0.81, 0.82, 0.81, and 0.81, respectively (Figure 3). Compared with the packing density study in concrete blocks ( 47 ), the packing density achieved in the present study was about 5.6% and 10% higher for NA and RCA mixtures, respectively. This improved packing in RCCP could be attributed to the difference in the compaction procedure, namely lower vibration duration and higher surcharge pressure. Further, Figure 3 shows that the denser aggregate skeleton could be produced with RCA similar to NA without affecting the compactness of RCCP mixtures. These findings demonstrate that 100% RCA can be used in RCCP applications.

Aggregate proportions and packing density for NA and RCA at different replacement levels.
Fresh Properties
Optimum Moisture Content
Typically, RCCP mixtures are designed at OMC to achieve maximum compactness, in accordance with ASTM D1557 ( 41 ). The moisture density plot for control and RCA mixtures at different moisture states is shown in Figure 4. As anticipated, RCA inclusions in the dry state entailed higher OMC of 20%, 30%, and 106% for coarse, fine, and total RCA mixtures, respectively, compared with the control mixture (Table 5). The increase in OMC was mainly rendered by the presence of microcracks and the porous nature of the AM, which increased the water absorption potential of dry RCA–RCCP mixtures ( 2 , 13 ). The individual replacement of coarse and fine RCAs exhibited a moderate increase in OMC (20%–30%), whereas total RCA inclusion showed a twofold increase in OMC owing to a collective water demand of coarse (5.3% and 6.1%) and fine (8.8%) RCAs (Figure 4a). In previous studies too, the replacement of 50% and 100% coarse RCA in a dry state demanded 28% to 37% and 33% to 77% higher OMC ( 7 , 48 ). However, this negative effect of RCA can be alleviated by using RCA in the SSD state as the old AM thirstiness could be satisfied by soaking the aggregates 24 hours before. Owing to the lower water absorption potential of RCA in the SSD state, OMC was found to be similar to that of the control mixture, as shown in Figure 4b.

Moisture density plot at different moisture states of RCA: (a) Dry state and (b) SSD state.
Fresh, Mechanical, and Durability Properties of RCCP Mixtures
Note: na = not applicable; RCA = recycled concrete aggregates; OMC = optimum moisture content; w/c ratio = water–cement ratio; SP = superplasticizer; SSD = saturated surface dry; MDD = maximum dry density; RCCP = roller-compacted concrete pavement. 1 kg/m3 = 0.062 lb/ft 3 ; 1 MPa = 145.038 pounds per square inch. Standard deviation for each result is given in parentheses.
Maximum Dry Density
The influence of moisture states of RCA on the compactness of RCCP mixtures is presented in Figure 4 and Table 5. As seen in Figure 4, the replacement of RCA lowers the density with the presence of its low-density AM, and consequently, a decrease in the density increases as the fractions of RCA increase ( 49 ). Including RCA in the dry state negatively affected the maximum compactness by 5.7%, 4.5%, and 14.5 % for coarse, fine, and total RCA-inclusive RCCP mixtures, respectively. Hosseinnezhad et al. ( 48 ) and Lopez-Uceda et al. ( 7 ) also reported that the inclusion of coarse RCA in the dry state at 50% and 100% levels lowered the MDD by 3%–3.7% and 5.2%, respectively. This is mainly attributed to the absorption of water during the mixing stage by the AM in RCA, thus reducing the free water content. This, in turn, affects aggregate rearrangement and compactability, resulting in lower MDD ( 50 ). A similar trend was also observed in the SSD state, where the reduction in MDD for coarse, fine, and total RCA mixtures was found to be 4.9%, 3.1%, and 11.7 %, respectively. However, compared with dry RCA, the use of RCA in the SSD state could slightly improve MDD. This enhancement is caused by the availability of initial moisture in the SSD state RCA. This results in more free water, which improves the compactability and density of RCCP mixtures.
Vebe Consistency Time and Vebe Density
The compactability of RCCP mixtures is generally determined by Vebe consistency time. The desirable Vebe time for better compactability is 20 to 40 s, in accordance with ACI 325.10R ( 51 ) and other research studies ( 52 , 53 ); a lower or higher Vebe time than the desirable range could result in heaving in front of rollers or poor compaction, which significantly affects the performance of RCCP. However, it should be noted that the desirable Vebe time shall consider the transit time of concrete and the application type. In this study, Vebe density and Vebe consistency time were evaluated by following procedure A in accordance with ASTM C1170 ( 37 ). It was observed that there was no formation of mortar ring until 60 s for control, dry, and SSD RCA mixtures, even though all the RCCP mixtures were designed at their OMC (Table 5). Meanwhile, efforts were made to improve the compactness of RCCP mixtures by including superplasticizers to achieve the optimal Vebe time of 20 to 40 s at dry and SSD states, as shown in Table 5.
The effect of moisture states and superplasticizers on the Vebe density is shown in Figure 5. The use of RCA in the dry state adversely affected the Vebe density by 6%, 15.5%, and 15.8% for coarse, fine, and total RCAs, respectively. This is attributed to the moisture absorption of RCA from cement paste during mixing, which decreases workability and adversely affects aggregate arrangement ( 22 ) (Figure 5a). Similar negative behavior was also observed in SSD RCA (Figure 5a); however, the Vebe density reduction rate was found to be lower, around 5.4%, 2.1%, and 5.7% for coarse, fine, and total RCA mixtures, respectively. These compactability issues could be counterbalanced by incorporating superplasticizers in the RCA–RCCP mixtures. The inclusion of superplasticizers slightly improved the Vebe density for coarse RCA and fine RCA mixtures in both dry and SSD states. For instance, a lower Vebe density reduction of 3.5% and 0.8% and 4.9% and 4.7% were observed for coarse RCA and total RCA mixtures in dry and SSD states, respectively. Interestingly, addition of superplasticizers showed a slight improvement in Vebe density for fine RCA mixtures by 1.5% in both dry and SSD states. A similar behavior was observed when conventional RCCP mixtures were produced with superplasticizers ( 57 ). This improvement could be attributed to better compaction facilitated by superplasticizers, thereby developing a denser aggregate structure for fine RCA mixtures (Figure 5b). Further, this might be linked to the homogeneous dispersion of cement paste and enhancement in the system fluidity as a result of the combined effects of the steric and electrostatic repulsion mechanism ( 55 , 56 ). Negatively charged ions (carboxylic group) adsorbed on positively charged cement grains result in electrostatic repulsion, whereas bulky side chains (long polymeric chains) of the absorbed polycarboxylate ether separate the cement grains apart ( 55 , 57 ), as shown in Figure 6.

Influence of moisture state and superplasticizer on: (a) Vebe density and (b) Vebe density ratio.

Dispersion mechanism of superplasticizer.
Mechanical Properties
Compressive Strength
The influence of moisture state and superplasticizer on the compressive strength of control and RCA–RCCP mixtures is presented in Table 5 and Figure 7. The control mixture achieved higher compressive strength at the designed moisture content than the RCA mixtures, except for the fine RCA in the SSD state with superplasticizer (Figure 7). Typically, RCA in the dry state continued to absorb the moisture from the mortar during the mixing process and further. To quantify the percentage of moisture lost during the mixing process, the water absorption characteristic of coarse RCA in the dry state for 12 h was recorded, as shown in Figure 8. It is apparent that during the mixing phase (<10 min.), dry state coarse RCA could absorb 49% of its total water absorption potential. This signifies that the initial 10 min. has a significant impact on the free water availability, thereby reducing the designed water–cement c ratio). Owing to the moisture/c ratio). Owing to the moisture absorption mechanism (Figure 9, a and
b

Influence of moisture state and superplasticizer on: (a) Compressive strength and (b) Compressive strength ratio.

Evolution of water absorption characteristics of coarse RCA.

Moisture transfer mechanism in RCA at dry (a and b) and SSD (c and d) states.
As the water demand for the AM in RCA was already satisfied in the SSD state, the effective w/c ratio is expected to be the same as the designed w/c ratio at the initial mixing stage. However, at later stages, the desorption of absorbed water from saturated RCA (SSD state) could have occurred toward the new cement mortar (internal curing) or ITZ near RCA, as shown in Figure 9, c and d . This might be mainly triggered by the difference in the moisture gradient as a result of cement hydration and water evaporation ( 50 , 59 ) (Figure 9, c and d ). Further, it could increase the effective w/c ratio near RCA, influence the ITZ between the RCA and new mortar, lower the mechanical bonding, and adversely affect the compressive strength ( 25 , 50 ) by 39%, 33%, and 40% for coarse, fine, and total RCA mixtures, respectively. Likewise, Poon et al. ( 25 ) observed that coarse RCA used in the SSD state exhibited a reduction in compressive strength of 19%. Nevertheless, the inclusion of RCA in the SSD state would provide a beneficial effect of lower shrinkage because of water release from RCA to meet the water demand of cement hydration ( 59 ).
In the case of addition of superplasticizers in dry and SSD states, a significant improvement in compressive strength was observed compared with the RCA in dry and SSD states, as shown in Figure 7. Interestingly, fine RCA with the inclusion of superplasticizers could achieve similar compressive strength as that of the control mixture in both dry and SSD states and could alleviate the detrimental effect of AM (Figure 7b). The inclusion of superplasticizers increase the cohesiveness of the concrete mixture owing to the homogeneous dispersion of cement particles, ease the compaction and lower the voids, and result in denser concrete mixtures ( 56 , 60 ). As a result of this, extra water needed to satisfy the water absorption of RCA could be mitigated, and better workability can be achieved ( 60 ). This combined effect of water reduction potential and lubrication effect of superplasticizers showed an improvement in the superplasticizer-inclusive RCCP mixtures in dry and SSD states (Figure 7a). A similar positive effect of the inclusion of superplasticizers was demonstrated by Hashemi et al. ( 57 ) for conventional RCCP mixtures. The authors ( 57 ) also concluded that the inclusion of superplasticizers could mitigate the urban heat island owing to the higher thermal conductivity, thereby rendering a cooler pavement. From the current findings, it can be concluded that 100% RCA with the inclusion of superplasticizers can be utilized for rural road construction; however, the same can be used for urban/high-volume roads by slightly increasing the binder content.
Indirect Tensile Strength
Consistent with compressive strength, the inclusion of RCA at different moisture states manifested a similar response in the indirect tensile strength. RCA inclusion in the dry state had the lowest tensile strength compared with the control mixture, whatever the aggregate replacement levels, as shown in Figure 10. This behavior is mainly because of the moisture loss during the mixing process, resulting in a lower degree of compaction for dry state RCA mixtures. Meanwhile, using RCA in the SSD state could slightly improve the compactability and tensile strength compared with the dry state RCA mix (Figure 10). Compared with the control mixture, a major decrease in the tensile strength of 28%, 17%, and 34% was observed for coarse, fine, and total RCA mixtures in the SSD state. This negative effect could be caused by the desorption characteristics of RCA in SSD state, thereby increasing the effective w/c ratio and porous ITZ near old and new mortar ( 61 ) (Figure 9). However, these adverse effects can be lessened by the inclusion of superplasticizers. The addition of superplasticizers in the dry state positively affected the tensile strength (Figure 10b). Surprisingly, the incorporation of superplasticizers in SSD state for coarse, fine, and total RCA mixtures manifested an emphatic increase in the indirect tensile by 47%, 28%, and 39%, respectively, compared with the control mixture. This behavior could be ascribed to the two governing mechanisms: (i) RCA possesses a higher angularity and surface texture than NA, which creates good bonding between the cement matrix and RCA and imparts resistance to separation of mortar during loading ( 62 ); and (ii) addition of superplasticizers exhibits a significant improvement in the degree of compaction because of lubrication effect provided by two repulsive actions (steric hindrance and electrostatic repulsion) (Figure 6). At 28 days of testing, the inclusion of superplasticizers has a more pronounced positive effect on tensile strength than the compressive strength of RCCP mixtures. These findings suggest that incorporating superplasticizers in the SSD state could reverse the negative effect of RCA and significantly improve the tensile strength of RCCP mixtures.

Effect of moisture state and superplasticizer on: (a) Indirect tensile strength and (b) Indirect tensile strength ratio.
Porosity and Water Absorption
The effect of incorporation of RCA and superplasticizer at different moisture states on the water absorption and porosity of RCCP mixtures are presented in Figure 11 and Table 5. As expected, the inclusion of RCA in the dry state significantly increased the porosity and water absorption for coarse, fine, and total RCA mixtures by 50% and 34%, 87.5% and 85%, and 180% and 160%, respectively. This is mainly caused by fractures and capillary pores in AM, which increases the water absorption and porosity of RCA in the dry state. Further, Leite and Monterio ( 32 ) observed that dry RCA would release air bubbles while absorbing water during the mixing stage, thus leading to the formation of macropores near ITZ, which increase the porosity. In the case of RCA in the SSD state, higher porosity and water absorption were observed owing to the increase in the effective w/c ratio rendered by internal curing of RCA (Figure 11). However, the inclusion of superplasticizers in both dry and SSD states RCA exhibited a slight improvement in compactability, which in turn reduced the voids and lowered the water absorption and porosity compared with dry and SSD states RCA mixtures. Although the inclusion of superplasticizers in dry and SSD states RCA manifested a positive effect on strength properties, it could not significantly improve water absorption and porosity characteristics because of RCA’s inherent property (AM).

Influence of moisture state and superplasticizer on: (a) Water absorption and (b) Porosity.
Conclusions
This study attempts to provide a framework for producing workable as well as durable RCCP mixtures containing 100% RCA. As the compactness of stiff mixtures such as RCCP has a direct relationship with the concrete performance, synergetic efforts were made to improve RCCP compactness by altering the moisture states of RCA, lubricating the mixtures through chemical admixtures, and enhancing the interlocking by blending the RCA following particle packing approaches. The inferences from this comprehensive investigation are as follows.
The blended RCA with the packing density approach achieved a similar packing density to the NA (0.81). This signifies that the RCA could also form a denser aggregate skeleton similar to the NA.
The use of RCA in the dry state negatively affected the compactness, compressive strength, and indirect tensile strength by 5.7% to 14.5%, 41% to 53.5%, and 30% to 42%, respectively. This was mainly rendered by the absorption of mixing water by RCA, which reduced the effective w/c ratio. This reduced w/c ratio negatively affected the water absorption and porosity of RCCP.
Incorporation of RCA in SSD state was also found to severely affect the compactness, and thus, the hardened state behavior—compressive strength and indirect tensile strength were found to be 33% to 40% and 30% to 42% lower than the RCCP containing NA. This could be a result of the desorption behavior of RCA in the SSD state, which increases the effective w/c ratio and forms the macropore near RCA.
Inclusion of superplasticizers in both dry and SSD states could alleviate the adverse effects of RCA. The lubrication and water reduction potential of superplasticizers tend to satisfy the water demand of RCA, subsequently improving the compactability, fresh density, and mechanical properties of RCA in both dry and SSD states. Interestingly, the inclusion of superplasticizers in the SSD state improved the tensile characteristics by 29% to 47% compared with the control mixture.
The use of RCA in SSD state with the conjunctive use of a superplasticizer could be recommended for low-volume rural roads considering it achieved the requisite fresh and mechanical properties.
Footnotes
Acknowledgements
The first author acknowledges the PMRF scholarship received from the Ministry of Education, Government of India. All the authors appreciate the funding received from the Indian Institute of Technology Madras, Chennai, India.
Author Contributions
The authors confirm contribution to the paper as follows: study conception and design: Surender Singh, M. Selvam; data collection: A. G. Anjana, M. Selvam; analysis and interpretation of results: M. Selvam, Surender Singh, Anjana A. G.; draft manuscript preparation: M. Selvam, Surender Singh. All authors reviewed the results and approved the final version of the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding was received from the Indian Institute of Technology Madras, Chennai, India, for the projects SB20210809CEMHRD008100 and CE1920900RFER008952.
