Properties of Concrete Prepared with Recycled Aggregates Treated by Bio-Deposition Adding Oxygen Release Compound

Recycled aggregates have high water absorption and crushing index. In order to improve the properties of recycled aggregates in concrete production, various treatments were used to modify the aggregates. In recent years, bio-deposition as a new treatment method of recycled aggregates was environmentally friendly. An improved method of bio-deposition was implemented to modify the properties of recycled mortar aggregates (RMA). O-bio-deposition is based on aerobic bacteria induced CaCO3 precipitation by respiration by varying the distance between the RMA and the bottom of the container and by adding an oxygen release compound to the culture solution that contains bacteria to promote the induction of CaCO3. First, the physical properties, including water absorption, crushing value, and apparent density, of the coarse RMA under different treatment methods were determined, and an o-bio-deposition treatment method was obtained. The fine RMA was treated and compared with the untreated RMA. Concretes were then prepared from the treated coarse RMA, and compressive strength and slump were determined. In addition, the effect of the o-bio-deposition treatment on the RMA surface and the micro-cracks of concretes were observed by scanning electron microscopy (SEM). It was found that the water absorption and crushing index of the coarse RMA treated by o-bio-deposition were reduced by 40.38 and 19.76% compared with untreated RMA, respectively. Regarding the concrete, the slump and the compressive strength (28 d) of concrete were increased by 115% and 25.3%, respectively compared with the untreated concrete.


Introduction
In China, with the rapid development of the construction industry, the amount of construction and demolition (C&D) waste produced increases. Approximately 20 billion tons of waste concrete were produced annually in 2017 [1]. From the perspective of sustainable development, recycling waste concrete to produce recycled concrete aggregate (RCA) can effectively reduce harm to the environment [2,3]. However, the attached mortar on the surface of recycled aggregates leads to higher water absorption and crushing value of RCA [4][5][6][7]. Verian et al. [8] reported that the water absorption of natural aggregates (NA) ranged from 0.34%-3.00% and RCA ranged from 0.50%-14.75%. Saravanshi et al. [6] reported that the crushing value for RCA is 33% higher than that of NA. Therefore, the compressive strength and durability of recycled aggregate concrete (RAC) prepared with RCA were negatively affected. [7,[9][10][11][12][13][14]; the compressive strength of RAC for 100% aggregates replaced by RCA can decrease by about 30% compared to concrete made with NA [5,8]. Currently, researchers environment. Wang et al. [30] designed several treatment methods based on spraying and immersion methods of bio-deposition. The results showed that the weight increased by 2% and the decrease in water absorption of the RCA was 10% after two immersion treatments. Qiu et al. [31] reported that the water absorption reduction of RCA using bio-deposition was 15% and the weight of RCA increased 0.3% under certain conditions of pH, temperature, bacterial concentration, and Ca 2+ concentration. However, ammonia produced by ureolytic bacteria not only caused corrosion of the steel bars but also polluted the environment [26]. Therefore, Wu et al. [34] used the bio-deposition method, which was based on Bacillus pseudofirmus (DSM8715) to induce CaCO 3 precipitation by respiration. The results showed that the water absorption and crushing index of the RCA treated by bio-deposition both decreased (10% and 15%, respectively).
In this study, regarding the selection of strains and the treatment method of bio-deposition, Zhang et al. [35] determined and screened a strain of H4 with the highest calcium precipitation activity (CPA) of 94.8% to self-heal concrete cracks. Furthermore, Zhang et al. [36] developed an oxygen-releasing compound (ORC) that contained calcium peroxide (CaO 2 ) and lactic acid that supplied oxygen (O 2 ) for the calcium precipitation of the H4 strain upon contact with water. In the presence of oxygen, strain H4 induced 50% more calcium precipitation than that obtained without oxygen. Therefore, in this paper, Bacillusal alcalophilus (H4) [35] was used to modify RMA. To supply a considerable amount of O 2 to the H4 strain to metabolize and produce CO 2 combined with Ca 2+ to induce CaCO 3, based on the above research, calcium peroxide (CaO 2 ) was innovatively added to the bacterial solution as an ORC [36]. For the first time, the effect of the distance between the aggregates and the bacteria in the container on the improvement was investigated by changing the position of the RMA in the container.
First, the effects of the concentration of CaO 2 in the culture solution on the respiration of aerobic strain H4 and the position of the RMA immersed in the container were considered; the optimum concentration of CaO 2 and the position of RMA immersion in the container were determined by testing the water absorption, apparent density, and crushing index of the coarse RMA under different treatment methods, and the optimal treatment method of o-bio-deposition was obtained. Then, fine RMA was treated with bio-deposition and o-bio-deposition, and the characteristics of the RMAs with different particle sizes were obtained and compared with those of the untreated RMA. Finally, concrete was prepared with treated and untreated coarse RMAs, respectively, and the slump of fresh concrete and the compressive strength of hardened concrete were obtained. Furthermore, SEM was used to observe the surface of the RMA and the micro-cracks of the concrete.

Recycled Mortar Aggregates
The RMA for the study was obtained from old mortar blocks. Based on the mixture ratio in GB/T 17671-1999 [37], an old cement mortar mixture was prepared with a sand to cement ratio of 2.5 and a water to cement ratio of 0.5. The mortar mixture was cast in steel molds. After 24 h, the specimens were demolded and maintained in a water curing tank at 20 ± 2 • C for 90 days. After curing, all specimens were removed from the tank and crushed using a jaw crusher, and the coarse aggregates (5-20 mm) and the fine aggregates (<5 mm) were obtained.

Cement
Ordinary Portland cement with a strength grade of P.O 42.5 was used, and the density of the cement was 3.16 g/cm 3 , whereas the specific surface area was 3519.5 cm 2 /g.

Bio-Deposition Treatment Method for RMA
The traditional way to improve RCA is to immerse RCA in the bottom of glassware filled with a culture solution-containing bacteria (called bio-deposition) (Figure 1a). In this study, CaO 2 was added as the ORC to provide O 2 and a small amount of calcium, and the lactic acid powder was added to regulate the pH to 10.0. Ca(OH) 2 is produced when CaO 2 releases oxygen, and Ca(OH) 2 is a micro-soluble substance; thus, the RMA was separated from CaO 2 by a screen. The concentration of CaO 2 added in the bacterial culture medium was set at a 5 g/L gradient, i.e., 5, 10, 15, 20, and 25 g/L.  Figure 1b). The RMA was immersed at 26 • C for 20 days. The fine RMA was treated using the o-bio-deposition method, and the fine RMA was obtained from the study of the coarse RMA.

Water Absorption
The test was performed according to the Chinese specification JGJ 52-2006 [38]. Bacillus alkalophilus H4 cannot metabolize in an environment above 60 • C [34], hence, the weight was obtained by drying at 50 • C.

Apparent Density
JGJ 52-2006 [38] was used to determine the apparent density of RMA. RMA was dried in an oven (50 ± 5 • C) to ensure bacterial activity in concrete preparation [33].

Crushing Index
The crushing index of the RMA was determined by the Chinese specification JGJ 52-2006 [38]. The coarse RMA with particle sizes of 10-20 mm was placed into a steel mold, and the surface was levelled and the mold was placed on the loading machine. The maximum load was 200 kN at a rate of 1 kN/s, and the load was retained for 5 s. After loading, the coarse RMA was poured out and determined (m 0 ). The weight of the residual RMA was determined by sieving with a 2.36-mm pore size sieve (m 1 ). The crushing index (Cc=C) of coarse RMA was calculated according to Equation (1).
Fine RMA was divided into four classes of different particle sizes: 5-2.5 mm, 2.5-1.25 mm, 1.25-0.63 mm, and 0.63-0.315 mm. The fine RMA was placed into the steel mold and loaded by a loading machine at a rate of 500 N/s to 25 kN and retained for 5 s. After loading, the fine RMA was sieved through 2.5, 1.25, 0.63, and 0.315-mm pore size sieves successively. The crushing index of each class was determined using Equation (1) (C = a 1 , a 2 , a 3 , and a 4 ); then, the total crushing index of fine RMA was determined using Equation (2) [34]. The values used were the average of three measurements. where: where: C sa = The total crushing index (%) was accurate to 0.1%. a 1 , a 2 , a 3 , and a 4 = the crushing index of fine RMA with different particle sizes of 2.50 mm, 1.25 mm, 0.63 mm, and 0.315 mm, respectively. c 1 , c 2 , c 3 , and c 4 = the corresponding residual weights (%)

Characterization of the Surface of RMA by SEM-EDS
Scanning electron microscopy (SEM) was used to observe that CaCO 3 that was induced by bacteria filled the pores and micro-cracks of RMA. Energy dispersive spectroscopy (EDS) analysis was conducted simultaneously to obtain the chemical composition of the particles found on the surface of RMA. The effect of the treatment on the micro-cracks in concrete was observed by SEM (Gemini SEM 300/VP ultra-high resolution field emission scanning electron microscope produced by Zeiss, Oberkochen, Germany).

Concrete Properties
The untreated and treated coarse RMAs were the only coarse aggregates used to make concrete with the mix shown in Table 1, and they were proportioned following the Chinese specification JGJ 55-2011, GB/T 25177-2010 [39,40]. The coarse RMA treated with the optimal treatment method of o-bio-deposition was used for making concrete (denoted as O-RMA/C). The coarse RMA treated by bio-deposition was used to make a control concrete (denoted as B-RMA/C) to examine the effect of the bio-deposition treatment. In addition, untreated coarse RMA was used to prepare concrete (denoted as U-RMA/C) as an extra reference. However, owing to the different water absorption of the aggregates, additional water was used to keep the saturated surface of the aggregates dry before mixing; thus, the actual total amount of water is shown in Table 1. The concrete mixture was poured in a 100 mm × 100 mm × 100 mm steel mold. After 48 h, the specimens were demolded and cured at a temperature of 20 ± 2 • C for 3, 7, 14, 28, and 56 days. The physical properties of the samples were then tested.

Slump
To compare the influence of the treated and untreated coarse RMA on the workability of the concrete mixture, the slump of the fresh concrete mixture was measured following the Chinese specification GB/T 50080-2016 [41]. The slump of the concrete mixture was determined to an accurate value of 1 mm.

Compressive Strength of Concrete
The compressive strength test was carried out on the side of the fractured block. The compressive strength was tested according to the Chinese specification GB/T 17671-1999 [37]. The compressive strength test was carried out at 3, 7, 14, 28, and 56 days.

The Optimal Treatment Method for the o-Bio-Deposition
According to the test mentioned in Section 2.2, the properties of the treated coarse RMA were obtained as shown in Figure 2. The results showed that the properties of the coarse RMA were significantly improved by adding CaO 2 solution to the bacterial solution and changing the immersion position of the coarse RMA compared to the traditional bio-deposition method. When the same concentration of CaO 2 was added to the culture solution, the modification of the properties of the coarse RMA were optimum in the middle of the immersion position; at the same immersion position, the water absorption and the crushing index of the aggregate first decreased and then increased with the concentration of CaO 2 , and the apparent density increased first and then decreased when the concentration of CaO 2 was 15 g/L. Thus, we observed the largest improvement in the properties of the coarse RMA.  This may be due to the fact that Bacillus alcalophilus H4 is an aerobic bacteria. There is a lack of O 2 in the water when treated by bio-deposition; therefore, most of the H4 strain was suspended on the surface of the culture solution to obtain O 2 , which caused the H4 strains to be far from the RMA, and CaCO 3 precipitation could not effectively accumulate on the surface of the RMA.
As for o-bio-deposition treatment, when the CaO 2 concentration ranges from 0 to 25 g/L, the addition of 15 g/L of CaO 2 as an oxygen release compound caused the dissolved oxygen concentration in the water to be the highest [42]. In the presence of oxygen, spores could maintain high metabolic activity, thus improving the ability of aerobic bacteria to produce CaCO 3 . In addition, CaO 2 reacted with water to produce not only oxygen but also Ca 2+ , which could be used as a supplement to Ca 2+ for bacteria to induce CaCO 3 [36]. CaO 2 could provide O 2 slowly, which caused the dissolved oxygen rich in the culture solution and the strain H4 was no longer suspended on the surface of the culture solution; instead, it may be because bacteria have a certain weight, the strain H4 accumulated in the middle of the container along with the microcirculation of oxygen in the culture solution. Therefore, when RMA was placed in the middle of the container, strain H4 was closer to the RMA; thus, CaCO 3 precipitation deposited to the surface of the RMA was more than that of bio-deposition, which affecte thed water absorption, and that in the pores mostly affected the crushing value. The weight of RMA increased in the same volume, thus the apparent density increased.
Comprehensively, when the coarse RMA was immersed in the middle and the concentration of the CaO 2 was 15 g/L, modification of the RMA was optimal. It was determined that the optimal treatment method was the o-bio-deposition processing method.
The fine RMA was treated with o-bio-deposition and bio-deposition. The water absorption, crushing value, and apparent density of the different particle RMAs are shown in Figure 3. The properties of B-RMA and O-RMA improved compared with U-RMA at all size fractions, and the modification of O-RMA was better than that of B-RMA. When the particle sizes of O-RMA were 20 mm and 5 mm, the water absorption of O-RMA was 40.4% and 25%, lower than that of U-RMA, respectively, and those of B-RMA were 28% and 15.8% lower than that of U-RMA, respectively. When the diameter of O-RMA was 20 mm and 5 mm, the crushing indices of O-RMA were 19.8% and 17.8% lower than that of U-RMA, respectively, and those of B-RMA were 8.7% and 7.8% lower than that of U-RMA, respectively. However, the improvement in the apparent density was relatively low. O-RMA and B-RMA decreased by 4.9%, 4.2%, 2.5%, and 2.1% respectively.
From Figure 3, it can be found that the improvement of o-bio-deposition on the properties of RMA at all particle sizes is better than that of bio-deposition. This is because o-bio-deposition not only improved CPA of aerobic bacteria by adding ORC, but also changes the distance between the bacteria and the RMA, so that CaCO 3 deposited on the surface of the RMA, filling the surface cracks of the RMA, and improving the properties of RMA. Kou et al. [25] reported using CO 2 to treat RMA and obtained similar results for water absorption when curing 24 h, that is, the water absorption was reduced by 38% for particle of 20 mm. Wu et al. [34] used bio-deposition to treat RCA using aerobic bacteria (DSM8715), the water absorption was reduced by 10% and 23% for particles of 20 mm and 5 mm, and the crushing value reduced by 15% and 12% for the particle sizes of 20 mm and 5 mm. Compared with previous studies, it can be found that the performance of RMA improved by o-bio-deposition was similar to that by CO 2 treatment. Moreover, the modification of bio-deposition using strain H4 on recycled aggregate is better than that of DSM8715. This is because different aerobic bacteria have different CPA (CPA of the strain H4 is higher than DSM8715) [35]. The surface of the untreated and treated RMA was observed by SEM, and the chemical composition of the deposited particles on the surface of RMA was determined by EDS (energy dispersive spectroscopy) (Figure 4). In the three specimens (Figure 4d), the particles were CaCO 3 as indicated by EDS analyses (Figure 4e-g). There was no CaCO 3 on the surface of the untreated RMA (Figure 4a). Calcite-type CaCO 3 crystals were produced on the surface, as well as cracks of B-RMA ( Figure 4b); however, they were not enough to fill the micro-cracks. Compared with B-RMA, several CaCO 3 crystals were observed on the surface and the cracks of the O-RMA to completely fill the micro-cracks (Figure 4c,d).    The improvement of compressive strength may be attributed to bio-deposition treatment. CaCO 3 produced by bacteria plays a positive role in improving the compressive strength, which improves the physical properties of B-RMA and O-RMA. As shown in from Figure 3, the water absorption of 20-mm O-RCA and B-RCA is 40.4% and 28% lower than that of U-RCA, and the crushing index of the O-RCA and B-RCA is 25% and 15.8% lower than that of the U-RCA, respectively. The physical properties of O-RMA are better than that of B-RMA. Figure 4 shows that the O-RMA was covered with more CaCO 3 particles than B-RMA. Fine CaCO 3 promotes a hydration reaction and enhances the ITZ (interfacial transition zone) [43]; thus, the compressive strength of O-RMA/C is higher than that of B-RMA/C. The compressive strength of the B-RMA/C and O-RMA/C increased more than that of the U-RMA/C from 3 to 7 d, and the increase was similar with the mass death of bacteria in the later period.

SEM Analysis of Micro-Cracks in Concretes
The micro-cracks of the concretes can be observed in Figure 7. There are evident cracks in the U-RMA/C (Figure 7a), and some hydration products near the cracks; however, there are not enough to fill and cover the cracks. There are also cracks in the interface transition zone of O-RMA/C ( Figure 7b); however, the width of the cracks is evidently narrow, and several hydration products near the cracks fill the cracks effectively. Hydration products exist in clusters near the cracks. These may be attributed to the fine calcium carbonate particles that can provide nucleation sites for the hydration products, thus promoting the hydration reaction for increasing the hydration products to fill the micro-cracks in the concrete and improve the compressive strength of concrete. [43].

Conflicts of Interest:
The authors declare no conflict of interest.