Performance Evaluation of Strengthening Recycled Coarse Aggregate in Cement Stabilized Mixture Base Layer of Pavement

)is paper attempts to study the effect of different strengthening methods on recycled coarse aggregate (RCA) reprocessed from construction and demolition (C&D) waste. And it aims to obtain the optimal RCA content in cement stabilized recycled mixture (CSRM) applied in the base layer of pavement. Firstly, three strengthening methods were used to strengthen the RCA. It was revealed that the sodium silicate solution is the best method in a strengthening concentration of 8% and strengthening for five hours. )en, the mechanical experiments under two cement dosages, including unconfined compressive strength, splitting strength, and compressive rebound modulus tests, were carried out. In each cement dosage, five different RCA contents (0%, 25%, 50%, 75%, and 100%) were considered. It was found that, at 5% of cement dosage, the strength of CSRM satisfied the requirements of the base layer of the expressway under heavy traffic load. In the case of 4% of cement dosage, the strength of CSRM which was beyond 50% of RCA content was failed to meet specification requirements. For the failed cases, by strengthening RCA in the best method obtained before, 7 d unconfined strength and crushing value of the CSRM with 50% RCA content could meet the requirements of the subbase of expressways and first-class highways in China; the 75% of the mixture meets the requirements of the second-class highways and below.


Introduction
e rapid development of China's urbanization has resulted in massive generation of C&D waste, and it represents more than 30% of the world's municipal solid waste, 2 billion tons in 2019 [1]. After separating, screening, and removal of impurities, converting C&D waste into RCA and then applying it into road engineering have become a hot issue at present. Many research studies show that C&D waste has been widely utilized as materials of pavement base [2,3] and subgrade fillers [4,5]. In addition, by the end of 2030, 47,000 kilometers of highways and 100,000 kilometers of national highways will be built in China [6]. e increasing demands of construction in highway will lead to significant increasing depletion of natural aggregate sources. Due to the strict restrictions on exploitation of natural aggregates, there is a growing gap between the demand and supply capacity of road sources materials. Utilization of RCA into the pavement construction industry contributes to improving the efficiency and sustainability by conservation of raw materials and reduction of the landfilled C&D waste amount and greenhouse gas emission [7][8][9][10].
Characterization studies of RCA have been conducted in recent years. Compared with natural coarse aggregate, RCA has higher crushing index, larger water absorption rate, and lower apparent density [11,12]. Microexperiments were carried out to analyze the reasons for these differences [13,14]. Some studies also have shown that the cement kilns, fly ash, etc. considerably help to improve its performance [15,16]. At the same time, in order to make better use of RCA, various enhancement methods have been proposed to improve the performance of RCA, including mechanical device [17], nanosilica and bacteria [18], waste basalt powder [19], and microbial carbonate precipitation treatment [20]. However, there were no experiments carried out to establish the relationship between the above strengthening methods and the mechanical properties of CSRM under the condition of different RCA contents and cement dosages.
At present, many scholars have carried out researches on the strength, stiffness, crack resistance, and shrinkage properties of CSRM by utilizing RCA as materials in the base layer of highway [21][22][23]. Researchers have found that, with the increase of RCA content, the mechanical properties will deteriorate [24][25][26]. Under the same cement dosage, after mixing with RCA, the compressive strength of CSRM will be reduced by 12% to 25% [27,28]. Butler [29] even concluded that it will be reduced by 30%. At the same time, the RCA will also have a certain impact on the flexural strength and unit wear loss of CSRM [11]. erefore, the natural coarse aggregate can be only partially replaced by RCA to meet the requirements of the pavement base. Mills-Beale suggested that the RCA content should not exceed 75% [30]. Under a single cement dosage, Poon and Chan [31] and Azam and Cameron [32] have carried out researches on the performance of RCA as a base or subbase layer of granular materials. Resilient modulus is also a mechanical performance index that can be studied [33,34]. Li et al. [35] concluded that the RCA content should not exceed 75% from the perspective of freeze-thaw splitting tensile strength, elastic modulus, and dynamic modulus.
In this paper, 3 different strengthening methods were compared to evaluate their effect on RCA. By comparing the changes in the water absorption rate and crushing value, the optimal strengthening method was selected.
en, under different cement dosages (3%, 4%), the mechanical properties of CSRM with different RCA contents (0%, 25%, 50%, 75%, and 100%) were studied. At cement dosage of 5%, the 7 d unconfined compressive strength satisfies the requirements of the base layer of the expressway under heavy traffic load. For these RCA contents at 4% of cement dosage that does not meet the requirements of the specification, the optimal strengthening method was applied to improve the mechanical performance of CSRM. Finally, by comprehensive evaluation, the recommended RCA content was proposed corresponding to different-class highways. e outcomes of this study are expected to provide a useful reference for pavement analysis and design, thus promoting the consumption of C&D waste.

Cement.
e cement used in this experiment was P.C32.5. Its basic performance indicators were measured, and the results are shown in Table 1. e standard value is obtained from China specification "General Portland Cement" [36].

Aggregate.
is study mainly focused on coarse aggregates with particle size of 4.75-26.5 mm. Due to a large number of impurities on the surface of the RCA reprocessed from C&D waste, it needs to be washed and dried before testing. RCA mainly includes recycled concrete aggregate and recycled brick.
ere are huge differences between them in flexural strength, permeability, and other properties [37,38]. erefore, the natural coarse aggregate and these two main ingredients of RCA were separated manually, as shown in Figure 1. According to the method of Chinese standard "Highway Engineering Aggregate Test Regulations" [39], the basic properties of the RCA with 3 different particle sizes were studied. And the test of water absorption rate was conducted by Volumetric flask method in this standard. e results are shown in Table 2.
According to Table 2, the water absorption rate of the recycled brick aggregate is higher than those of the natural aggregate and recycled concrete aggregate. With the increasing in brick aggregate particle size, its water absorption rate slowly decreases.
e crushing values of these two main ingredients of RCA can meet the requirements of the base and subbase of China's standard "Technical Rules for Construction of Road Pavement Base Courses" [40]. But the values of the concrete aggregate and brick aggregate are 25.47% and 39.62% greater than those of natural aggregates, respectively. e difference between these two main ingredients and natural aggregates is mainly because of the differences in their surface state.
ere are some mortars attached on the surface of recycled concrete aggregates and recycled brick aggregates, which results in more pores on the surface of the RCA. Compared with recycled concrete aggregates, there are more mortars attached to recycled brick aggregates.
us, the water absorption rate and crushing value of recycled brick aggregates are larger than those of recycled concrete aggregates [41,42]. e indexes of fine aggregate are shown in Table 3. As can be seen from Table 3, the apparent density and water absorption rate of recycled fine aggregate are, respectively, smaller and higher than those of natural fine aggregate.
e water absorption rate of recycled fine aggregate is 75.65% greater than that of natural fine aggregate.
is is mainly because the recycled fine aggregate is composed of some stones in the crushing process and unhydrated mortar or brick aggregate.

Grading Design.
e grading of the mixture adopted the median value of the skeleton dense grading range specified in China's standard [40], as shown in Figure 2.
In order to determine the optimal moisture content (OMC) and maximum dry density (MDD) of the CSRM, the compaction test was carried out. Since the recycled brick aggregate accounted for 10% in RCA, the influence of the recycled brick aggregate on the properties of CSRM could not be ignored. e ratio between recycled concrete and recycled brick aggregate was approximately 9 : 1. en, the natural coarse aggregate was replaced by RCA in this ratio to carry out the compaction experiment. e RCA replacement rates are 0%, 25%, 50%, 75%, and 100%. In addition, two different cement dosages (4%, 5%) were considered. e results are shown in Figure 3. e following information can be obtained through Figure 3: (1) As the content of RCA increases, the MDD of the CSRM decreases and its OMC increases, regardless of the cement dosage. Taking the cement dosage of 4% as an example, when the content of RCA increases from 0% to 100%, the MDD decreases from 2.32 g/cm 3 to 1.90 g/cm 3 . e OMC increases from 4.11% to 10.52%. (2) e MDD and the OMC of the CSRM, respectively, increase and decrease with the increase of cement dosage regardless of the RCA contents. is is mainly because more water is needed to hydrate the CSRM with the increase of cement dosage. During the compaction, hydrated products will gradually fill the voids of the CSRM, which contributes to the increase of the MDD.

Research Content.
is article mainly studies different strengthening methods for strengthening the RCA and mechanical properties of CSRM with different cement dosages. In each cement dosage (4%, 5%), five different RCA contents (0%, 25%, 50%, 75%, and 100%) were considered. e research contents are as follows: (1) Basic material performance tests were conducted to compare the performance of RCA and natural aggregates. e main performance indicators were apparent density, water absorption, crushing value, etc.
(2) ree different strengthening methods were adopted to strengthen the RCA. Different working conditions were set in each method.    ese experiments included unconfined compressive strength, splitting strength, and compressive rebound modulus tests. By analyzing the influence of cement dosage and the RCA content on the performance of CSRM, a recommended range of RCA was obtained in road engineering. (5) For the content of RCA that does not meet the requirements of the specification, the best strengthening method obtained in the previous tests was utilized to strengthen the RCA. By comparing three mechanical properties before and after strengthening, the recommended content of RCA after strengthening was obtained.

LA Abrasion Test.
LA abrasion test is a physical strengthening method, designed to make RCA collide with steel balls. After friction, the surface of the mortar was shedding, thus obtaining reinforced aggregates. Turns of the LA abrasion test (200, 400, and 800) and steel balls (4, 8, and 12) were used as two variables in this experiment. To carry out this experiment, 5 kg each of recycled concrete aggregate and recycled brick aggregate was utilized. e water absorption rate and crushing value of the RCA after strengthening are shown in Figure 4.
It can be seen from Figure 4 that the water absorption rate and crushing value take the minimum value at 400r. From 0 to 400r, the water absorption rate and crushing value of the RCA decrease. e water absorption rates of recycled concrete aggregate and recycled brick aggregate decrease by 1.21%, and 0.72%, respectively. e crushing values related to these materials decrease by 3.7% and 4.36%, respectively. When the cycles increase from 400 to 600r, they show an upward trend. For recycled concrete aggregate, the water absorption rate and crushing value averagely increase by 0.07% and 0.4%, respectively. For recycled brick aggregate, they sharply increase by 0.31% and 0.86%, respectively. is fact is because microcracks on the surface of RCA arise as the number of turns increases, and this phenomenon increases its water absorption and crushing values. But for recycled brick aggregate, it causes cracks and even damage. As shown in Figure 4(b), by colliding with 12 steel balls, the water absorption rate and crushing value increase from 400 to 600r. e greater the number of steel balls, the greater the damage to the recycled brick aggregate. at is the reason why the greater the number of steel balls for recycled brick aggregate, the greater the increase in water absorption rate. It can be easily concluded that the 12 steel balls and 800r are the most proper condition.

Sodium Silicate Solution.
e colloidal particle structure in the sodium silicate solution can be expressed by Figure 5, and the hydrolysis reaction can be briefly summarized as (1). Sodium silicate solution can react with calcium hydroxide to form C-S-H gel as shown in (2): After mixing the sodium silicate solution and RCA, the surface of aggregate is coated with a thin layer of sodium silicate film, that is, the adhesive film [43]. Adjacent aggregates are connected by an adhesive film to  Advances in Civil Engineering form an adhesive bridge ( Figure 6). en, the adhesive strength is established through the gelation process of sodium silicate.
e basic indicators of sodium silicate solution are shown in Table 4. e mass percentage of the sodium silicate solution was 41.64%. By adding water, it was prepared at 5%. To ensure that the solution can fully submerge the aggregate, the sorted recycled concrete aggregate and recycled brick aggregate were poured into the sodium silicate solution and stirred evenly. At the soaking time of 2h, 5h, and 10h, the recycled concrete aggregate and the brick aggregate were taken out and filtered out. en, they were spread on the enamel tray to naturally dry in the room. e water absorption rate and crushing value of the dried aggregates are measured. e above operation was repeated with 8% and 12% sodium silicate solution. e results are shown in Figure 7.
It can be seen from Figure 7 that as the strengthening time increases, the water absorption rate of two main compositions of RCA decreases at a concentration of 5%. As the substances in the sodium silicate solution react with the old mortar on the surface of the RCA, the hydrated silica colloid fills the pores and cracks on the surface of the aggregate. e change in crushing value also meets this law.

Advances in Civil Engineering
For recycled concrete aggregate, its water absorption rate and crushing value, respectively, decrease 1.21% and 1.5% when soaking from 0 to 10 h; for recycled brick aggregate, these two indexes decrease 3.99% and 4.6%. When the concentration is 8% and 12%, the higher the concentration of the sodium silicate solution, the more intense the reaction with the mortar on the surface of the RA, and more C-S-H gel will also be produced. erefore, the water absorption rate and crushing value reach the maximum in the 5 h of strengthening time. As strengthening time increases, the sodium silicate solution will form a coating on the surface of the RCA and the layer thickness gradually increases, which in turn causes the decrease of its strength [44]. In terms of water absorption rate, it is when the concentration is 8% and strengthening is five hours that the value reaches the minimum. e crushing value associated with 12% sodium silicate solution is slightly lower than that of 8%. But at 12%, the crushing value is more sensitive. With the increase of strengthening time, its value decreases.

Titanate Coupling Agent.
e titanate coupling agent used in this experiment was TM-200S. e mechanism of titanate coupling agent has always been an issue to research.
ere are many coupling theories about the coupling agent [45][46][47]. e principle of chemical bond is the most traditional and classical theory in the theory of coupling between the titanate coupling agent and inorganic filler [48]. It was proposed that titanium-derived coupling agents react with free protons at the inorganic interface to form organic monomolecular layers on the inorganic surface. e compatibilization of the inorganic/organic phase resulted in new composite property, catalysis, adhesion, and rheology performance. Its structural composition and action mechanism are shown in Figures 8 and 9.
In this figure, 1 ≤ M ≤ 4, M + N ≤ 6; R is short chain alkane group; R′ is long chain alkane group; X is C, N, P, S, etc.; Y is group of hydroxy, amino, epoxy, double bond, etc.
To carry out the experiments, 2 kg of 10 mm and 20 mm recycled brick aggregate and recycled concrete aggregate was poured into titanate coupling agent solution (1%, 2%, and 3%). e soaking time was 1 h, 2 h, 3 h, 4 h, and 5 h. After reaching the specific strengthening time, they were taken out and dried indoors. Finally, the water absorption rate and crushing value were measured according to the method in the specification. e results are shown in Figure 10.
It can be seen from Figure 10 that as the strengthening time increases, the water absorption rate and crushing value of the RCA also gradually decrease. e water absorption rate and crushing value rapidly decrease within 1-3 h of strengthening time. After 3 h, the water content and crush value tend to be stable. As the concentration of the titanate coupling agent increases from 1% to 2%, a hydrophobic layer can be formed on the surface of the RCA after the hydrolysis of the coupling agent. e number of pores on the surface of the aggregate decreases, thus strengthening RCA. By chemical reaction with the inorganic filler, the surface of the filler is coupled and cross-linked with the polymer matrix to combine two substances with different properties. It can form a monomolecular covering film on the surface of the filler, which contributes to improving its inherent hydrophilic properties and preventing moisture from entering into the aggregate. Lu et al. [49] also applied this principle to the surface modification of microfibrillated cellulose. e optimal concentration of the titanate coupling agent is determined by the RCA particle size and conditions of the surface functional group. e finer the filler particle size and the larger the surface area, the greater the amount of coupling agent used. When the strengthening time is 4 h and the concentration of titanate coupling agent is 2%, the effect of enhancement is the best.
By comprehensive comparison, the best working conditions of each strengthening methods are listed as follows: (   Advances in Civil Engineering e results of the above three strengthening methods are shown in Figure 11.
rough Figure 11, it can be clearly seen that the sodium silicate solution is the most ideal strengthening method. In this way, the water absorption rate and crushing value of the RCA are less than those of the other two strengthening methods. Specifically, the crushing value of RCA is close to that of natural aggregate, and the water absorption rate has also been greatly improved. Regarding the crushing value, the standard requires that the subbase layer of the first-class highways and expressway should not be higher than 26%. By strengthening with a 26% titanate coupling agent solution for 4 h, the crushing value of the recycled asphalt mixture can reach 24.6%, which meets the standard requirements [40]. For recycled concrete aggregate, the strengthening effect of the LA abrasion tester is better than the titanate coupling agent. For recycled brick aggregate, the effect of the titanate coupling agent is better than the LA abrasion tester. Based on the above analysis, the best strengthening method is strengthening with a concentration of 8% and strengthening for 5 h.

Unconfined Compressive Strength.
e unconfined compressive strength tests were carried out according to T0805-2009 test method in the standard [50]. e results are shown in Figure 12.
In this figure, K indicates the RCA content. By analyzing Figure 12, it is possible to conclude that, during the curing period of the CSRM from 7 d to 90 d, the strength growth rate from 7 d to 28 d is faster than that from 28 d to 90 d. Taking the dosage of 4% cement as an example, when the content of RCA increased from 0% to 100%, the unconfined compressive strength from 7 d to 28 d increases by an average of 35.02% compared to the stage from 28 d to 90 d. It is confirmed that the hydration reaction of cement mainly occurs in the early stage.

Advances in Civil Engineering
Moreover, when the cement dosage is 4% and the age is 7 d, the unconfined compressive strength corresponding to the 25% RCA content is the maximum, 4.16 MPa. Compared with the 90 d unconfined compressive strength of cement stabilized natural mixture, that parameter related to CSRM increases by 5.33% in the 25% of RCA content, while it decreases by 10.22%, 14.89%, and 19.33% under the condition of 50%, 75%, and 100% RCA content, respectively. Studies have shown that old mortar is contained in a suitable content of RCA. And its rough surface makes benefits to the binding of aggregates and cement. For the cement dosage of 4%,  when the curing age is 28 d and 90 d, the unconfined compressive strength of the CSRM firstly increases and then decreases with the increase of the RCA content. As the curing age grows, there is some unreleased bound water in the RCA. e water is released to promote the progress of hydration. When the content of RCA exceeds a certain limit, the mechanical properties of CSRM deteriorate. Some studies also obtained the same conclusion from the perspective of microanalysis and macroperformance [42,51]. Unlike the trend of unconfined compressive strength in cement dosage of 4%, the strength of the mixture decreases with the cement dosage of 5% when the RCA content increases. As the cement dosage increases, the disadvantage of low strength with RCA becomes more and more obvious, which makes the overall strength of the CSRM decrease.

Splitting Strength.
e results of the splitting tests are shown in Figure 13.
It can be seen from Figure 13 that the changing trend of the splitting strength of CSRM is roughly the same as that of unconfined compressive strength. is fact is because the splitting strength under the same material has a certain correlation with the unconfined compressive strength. At a cement dosage of 4%, the splitting strength increases firstly and then decreases with the increase of RCA content. e maximum splitting strength appears at 25% of the RCA content in all ages. At a certain content of RCA, the characteristics of the old mortar attached to the RCA and its rough surface play a greater advantage than the shortcomings of its strength. When exceeding this ratio, as the RCA content increases, the splitting strength decreases. Compared with the 90 d splitting strength of cement stabilized natural mixture, its value decreases by 5.36%, 16.07%, and 26.79% under Advances in Civil Engineering the condition of 50%, 75%, and 100% of RCA content. When the cement dosage is 5%, as the RCA content increases, the strength of the mixture gradually decreases, which is caused by the defects of its strength. e 90 d average strength of CSRM with a 5% cement dosage is 0.622 MPa, which is 0.112 MPa higher than that of 4% cement dosage, while the 7 d average strength with 5% cement dosage is only 0.06 MPa larger than that of 4%. As the curing age increases, the advantage in the strength of the high cement dosage is gradually revealed.

Compressive Rebound Modulus.
e results of the compressive rebound modulus tests are shown in Figure 14.
It can be noticed from Figure 14 that the change law of the compressive rebound modulus is the same as the two mechanical properties measured before. Compared with the 5% cement dosage, the strength is lower at the 4% cement dosage. Taking the cement stabilized natural mixture as an example, the 90 d compressive rebound strength of 5% cement dosage is 291.67 MPa greater than that of 4%. At 5% cement dosage, the 7 d unconfined compressive strength satisfies the requirements of the expressway base layer under heavy traffic load. Except for the 0% and 25% of RCA content, the value of splitting strength and compressive rebound modulus of 4% of cement dosage fail to meet the requirements of China standard "Highway Asphalt Pavement Design Criterion" [52] in the other content. In the existing research, there is no scientific basis to support the use of RCA to replace 100% of natural aggregates in the pavement base layer. erefore, in this paper, the CSRM with two different RCA contents (50%, 75%) at 4% cement dosage were strengthened.

Comprehensive Evaluation.
As can be seen from the strengthening method section in this study, 8% sodium silicate solution and strengthening for 5 h are the best strengthening method. Firstly, by strengthening the RCA and then comparing the mechanical properties of CSRM with 50% and 75% RCA contents, the range of the RCA content strengthening with sodium silicate solution was recommended.
e unconfined compressive strength, splitting strength, and compressive rebound modulus were measured at 50% RCA content before and after strengthening. e results are shown in Figure 15.
It can be seen from Figure 15  e results of the mechanical performance of CSRM at 75% content of RA are shown in Figure 16.
It can be seen from Figure 16   unconfined compressive strength and splitting strength. ese two mechanical performance indexes increase by more than 20%. e 7 d unconfined strength has also met the requirements of the subbase for second-class highway subjected to heavy traffic in China specification "Highway Asphalt Pavement Design Criterion" [52].

Conclusions
As a promising material in the pavement base, RCA play an increasingly significant role in the consumption of C&D waste and promoting sustainability of the construction industry. In this study, laboratory experiments were carried out to study the effect of different strengthening methods on RCA and mechanical properties of CSRM. e following conclusions can be summarized from the previous findings and observation: (1) Compared with natural aggregates, the RCA has a higher water absorption rate, a smaller apparent density, and a larger crushing value. Especially for recycled brick aggregate, its water absorption rate is dozens of times of natural aggregate. (2) Comparing the water absorption rate and crushing value before and after strengthening, the best strengthening way is obtained among different working conditions of the three strengthening methods. e sodium silicate solution is the best method in a strengthening concentration of 8% and strengthening duration of five hours.
(3) When the cement dosage is 4%, with the increase of the RCA content, the unconfined compressive strength, splitting strength, and compressive rebound modulus of the CSRM firstly increase and then decrease. When the RCA content is 25%, the mechanical properties of CSRM are the best. It can meet the strength requirements of the base and subbase layer of highway and first-class expressway. When the cement dosage is 5%, all these mechanical properties all show a decreasing trend when the RCA content increases. (4) Strengthening treatment was carried out for CSRM at 4% of the cement dosage. When it comes to 50% and 75% RCA content, the mixture does not meet the requirements of the specifications. By comparing the value of the three mechanical properties before and after strengthening, it was found that the strength growth rate of the 75% CSRM after strengthening is significantly higher than that of 50%. After strengthening treatment, the strength of CSRM with 50% RCA content has met the strength requirements of the base and subbase layer of highway and firstclass expressway. e 75% of RCA content can meet the strength requirements of the subbase of secondclass highways or inferior highways.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.