Eco-friendly concrete: Combining treated wastewater and recycled aggregates

Concrete is the most widely consumed construction material. Its extensive usage can be attributed to its exceptional strength, long-lasting durability, ease of mixing, and versatility in shaping. Concrete is a composite material formed by combining cement with natural resources such as water, sand, and gravel. However, despite water and aggregates being renewable resources over the long term, the growing population and urban development have led to alarming levels of concrete consumption.

The concrete industry stands as the largest water consumer (Asadollahfardi et al. 2016). In 2012, concrete manufacturing accounted for approximately 1.7% of global water consumption (Miller et al. 2018). Water plays a crucial role not only in the mixing and curing processes of concrete but also in activities like cleaning aggregates and mixer trucks. Yahyaei et al. (2020) state that the production of 1 m³ of concrete requires a minimum of 800 L of water, with 500 L designated for mixing and the remaining 300 L used for cleaning mixer trucks. Moreover, to maintain the physical and mechanical properties of concrete, it is common practice to employ water that is safe for drinking.

However, the issue of water scarcity is a stark reality in various regions worldwide. Consequently, the provision of potable water for concrete manufacturing has emerged as a significant concern in modern societies, prompting the exploration of alternative water sources that do not have drinking water quality. Municipal and industrial wastewater, after undergoing appropriate treatment, have gained recognition as viable alternatives due to their composition of 99.9% water and 0.01% solids. Such treated wastewater can achieve suitable quality for multiple reuse purposes, offering a promising solution. In addition, the continuous generation of wastewater, particularly in densely populated urban areas, further highlights its relevance as a significant water source. Consequently, numerous researchers have explored the potential of utilizing effluents derived from various wastewater treatment processes as a substitute for potable water in concrete production. Their investigations have revealed that it is indeed feasible to produce concrete with acceptable properties using such wastewater, as discussed in Almeida & Tonetti (2022).

While certain authors (Tay & Yip 1987; Tay 1989; Al-Joulani 2019; Duarte et al. 2019; Tonetti et al. 2019; Bouaich et al. 2022) have documented positive outcomes when potable water was substituted with wastewater, the majority of extant studies employing wastewater have indicated reductions in compressive strength. Nonetheless, these reductions remain within the 10% threshold prescribed by relevant standards (Asadollahfardi et al. 2016; Asadollahfardi & Mahdavi 2018; Brandão et al. 2019; Catanzaro et al. 2019; Saxena & Tembhurkar 2019; Hassani et al. 2020a; Yahyaei et al. 2020).

Another significant concern within the construction industry is related to the management of construction and demolition waste (CDW). Globally, the annual generation of CDW exceeds 10 billion tons, with the United States, the European Union, and China being the primary contributors, producing 700 million tons, 900 million tons, and 2,300 million tons, respectively (Cantero et al. 2020). Furthermore, the process of urbanization has made it increasingly challenging and costly to obtain natural sand and gravel due to the distance between consumers and producers, as well as the depletion of natural aggregate sources. Consequently, the utilization of CDW as recycled aggregates has emerged not only as an alternative to mitigate the consumption of natural aggregates but also as one of the most effective means to reduce the environmental impact caused by the civil construction industry (Silva et al. 2015). As a result, the incorporation of CDW in concrete manufacturing has become the subject of extensive study by numerous researchers in recent years (Martínez-Lage et al. 2012; Beltrán et al. 2014; Etxeberria & Vegas 2015; Silva et al. 2018, 2019; Cantero et al. 2020; Plaza et al. 2021).

With the increasing dissemination of this practice, CDW was classified into two categories: concrete recycled aggregates (CRA) and mixed recycled aggregates (MRA). Compared with CRA, the MRA inherently exhibits higher variability, reduced strength, and increased water absorption due to the presence of crushed mortar, bricks, and tiles (Etxeberria & Gonzalez-Corominas 2018; Martínez-Lage et al. 2020), and is consequently less frequently considered for concrete manufacturing activities.

However, it is important to note that depending on the intended application of the concrete (i.e. paving, blocks, filling, non-structural or structural use, etc.) and the desired performance requirements, it is indeed possible to incorporate recycled materials and achieve composites of excellent quality with satisfactory mechanical performance, all while reducing the consumption of natural resources.

Hence, the use of treated wastewater and recycled aggregates in concrete manufacturing becomes particularly relevant to the concrete industry when considering the escalating challenges posed by water and aggregate scarcity, and the disposal of construction debris in urban areas. However, it is important to note that the combined utilization of treated wastewater and recycled aggregates in concrete production is still a relatively incipient field of study, with limited research available on this subject and its impact on the physical and mechanical properties of concrete (Elchalakani & Elgaali 2012; Ramírez-Tenjhay et al. 2016; Raza et al. 2020, 2021; Ahmed et al. 2021). Therefore, the objective of the present study is to produce concrete by combining both materials and evaluating their performance (physical and mechanical properties), aiming to encourage their utilization if positive outcomes are found.

The mixed recycled aggregate, comprising both fine and coarse portions, was sourced from a recycling plant (SBR Reciclagem) located in Valinhos (Brazil). In addition, quartz sand and granite gravel were utilized as the natural aggregates in the study. All aggregates underwent thorough characterization with their respective properties assessed in accordance with the relevant Brazilian standards as detailed in Table 1.

A high early-strength cement, designed to achieve a 28-day expected strength of 59 MPa, was utilized in the study. The fresh properties of concrete, such as setting time and consistency, were assessed in accordance with the relevant Brazilian standards, as detailed in Table 1.

The control group (C0) exclusively employed natural aggregates and drinking water in its composition. In the CX-E family, treated wastewater completely replaced the drinking water component, while the mixed recycled gravel substituted X% (20, 40, 60, 80, and 100%) of the natural coarse aggregate. Within the C100/Y-E family, both the gravel and drinking water were entirely replaced by their respective alternative materials. Furthermore, the finely mixed recycled aggregate replaced Y% (10, 20, 30, 40, and 100%) of the natural sand in this family.

The mix design for 1 m³ of concrete for each specific mixture is provided in Table 2. A total of 319 cylindrical samples were carefully molded and subsequently subjected to a standardized curing process utilizing the same type of water employed during the mixing stage. For each mix design, a total of 29 samples were cast, of which 15 were utilized for compressive strength testing at different time intervals (7, 28, and 90 days). Furthermore, 10 samples were allocated for assessing splitting tensile strength at 7 and 28 days, while an additional four samples were employed for determining density, water absorption, and air-void ratio. All tests conducted, both in the fresh and hardened states, were conducted according to the relevant Brazilian standards as outlined in Table 1. Statistical analysis was performed utilizing analysis of variance (ANOVA), followed by the post hoc Dunnett's bilateral test. A significance level of p < 0.05 was considered to denote statistical significance.

The water used in concrete manufacturing is typically required to be free from impurities in order to prevent any alterations in cement hydration and subsequent changes in concrete properties. Consequently, drinking water is commonly considered the safest and most widely adopted option. Nonetheless, recognizing the issue of water scarcity in various countries, certain international standards provide recommendations regarding the utilization of alternative water sources, including specific guidelines outlining acceptable physical and chemical limits, as well as permissible variations in concrete properties such as compressive strength and setting time. Those recommendations are based on the threshold levels of substances that can have harmful effects on concrete. It is well known that chloride attacks the passive layer protecting steel reinforcement, leading to corrosion. Additionally, contact between concrete and high levels of sulfate can cause fissures and concrete disaggregation. Therefore, it is important to define safe concentrations of certain chemicals in the mixing water of concrete to ensure the production of a high-quality composite.

As demonstrated in Table 3, the effluent obtained from ETE Barão Geraldo exhibited physical and chemical properties that fell below the recommended limits outlined by Brazilian, British, and North American standards for concrete manufacturing, raising no concern about its influence over concrete strength and setting time. It is worth noting that the presence of total solids, which could potentially impact the mechanical and physical performance of cementitious composites, does not raise concerns in this case. Although parameters, such as zinc, lead, phosphate, and sulfate, were not evaluated in this study due to technical limitations, Duarte et al. (2019) conducted an analysis of the same effluent from ETE Barão Geraldo and determined that these particular parameters remained well below the maximum levels suggested by the standards (BS 2002; ABNT 2009b; ASTM 2012). Albeit drinking water does not require testing prior to its use in concrete manufacturing, Table 4 presents its physicochemical properties to facilitate a comparison between the different types of water utilized in the study.

Other authors using secondary treated effluent also obtained quality parameters within the standard limits, such as Al-Ghusain & Terro (2003), Terro & Al-Ghusain (2003), Asadollahfardi et al. (2016), Ghrair & Al-Mashaqbeh (2016), Oliveira et al. (2016), Saxena & Tembhurkar (2019), Catanzaro et al. (2019), Tonetti et al. (2019), Hassani et al. (2020a), Ahmed et al. (2021) and Arooj et al. (2021). Although it is important to highlight that both the British and Brazilian standards do not advocate in favor of using treated wastewater to produce concrete, the North American standard does not explicitly expose any prohibition. However, due to the satisfactory results obtained by the present and the aforementioned authors, we should question these rejections and encourage future research since wastewater is a continuum source of water and reuse is becoming a necessity.

The composition of the recycled aggregate, determined through visual segregation of the coarse fraction, was found to be as follows: 80.06% concrete and mortar, 8.32% asphalt, 8.66% bricks, 2.60% tiles, and 0.36% floating particles. Given that both the coarse and fine aggregates originate from the same source, and that the fine fraction was obtained by the simple crushing of the coarse fraction, it is considered that both fractions have a similar composition. In accordance with the classification proposed by Agrela et al. (2011), both fractions are considered as mixed recycled aggregates.

The cement paste prepared using drinking water and treated wastewater achieved the normal consistency with the same water-to-cement ratio of 0.32, consistent with the findings reported by Lee et al. (2001). The initial and final setting times using drinking water were 2 h 49 min and 3 h 47 min, respectively. When using treated wastewater, the initial and final setting times were 2 h 57 min and 3 h 50 min, respectively. Then, the incorporation of treated wastewater resulted in an 8-min delay in the initial setting time and a 3-min delay in the final setting time. However, it is important to note that these delays remained within the acceptable limits specified by the American (ASTM 2012), British (BS 2002), and Brazilian (ABNT 2009b) standards.

Similar findings were reported by Tay (1989), Lee et al. (2001), Ghrair & Al-Mashaqbeh (2016), Brandão et al. (2019), Catanzaro et al. (2019), and Ahmed et al. (2021), which corroborate the observed delays in setting times. These delays were expected due to the presence of residual materials, which can inhibit the hydration of C₃A and subsequently impede the development of ettringite, resulting in prolonged setting times (Asadollahfardi et al. 2016; Peighambarzadeh et al. 2020; Abushanab & Alnahhal 2021; Bouaich et al. 2022). Comparing wastewater from different treatment levels, Al-Ghusain & Terro (2003), Ghrair & Al-Mashaqbeh (2016), and Terro & Al-Ghusain (2003) observed that the lower the treatment level, the greater the delay in cement paste setting time. However, Bouaich et al. (2022) emphasize that these delays can be advantageous in hot weather conditions and for concreting projects involving long distances and large volumes.

The decrease observed when using treated wastewater can likely be attributed to the spongy surface of residual sludge particles, which have a tendency to absorb water. This absorption reduces the availability of free water and negatively affects the consistency of the mixture (Meena & Luhar 2019; Bouaich et al. 2022). Similarly, the use of recycled aggregates, characterized by their inherently higher porosity, rough and irregular surfaces, can result in the absorption of free water to fill the pores, contributing to a decrease in consistency (Agrela et al. 2011; Lima et al. 2013; Silva et al. 2014; Cartuxo et al. 2015). The dry consistency, in turn, affects the effective compaction of samples, leading to the formation of voids. Consequently, this results in concrete with reduced resistance and durability. Although it is common practice to pre-saturate aggregates or use admixtures or additional water to mitigate workability loss, the present study did not employ any of these methods in order to evaluate the natural variation in consistency resulting from the combined use of alternative materials.

It is noteworthy that the utilization of treated wastewater and increasing proportions of recycled aggregate led to a noticeable reduction in consistency. However, it is important to emphasize that compaction difficulties were observed exclusively in the samples of mix C100/100-E. This issue is attributed to the complete replacement with recycled aggregates, which have significantly higher water absorption compared with natural aggregates (see Table 3). This high absorption reduces the amount of free water available, resulting in an excessively dry consistency.

When compared with the control group (refer to Figure 5), the mean compressive strength of the CX-E mixes exhibited a range of +15.2 to −10.0%, +14.4 to −17.2%, and −3.3 to −30.8% at 7, 28, and 90 days, respectively. On the other hand, the C100/Y-E mixes demonstrated higher variations, ranging from −23.7 to −38.5%, −24.0 to −55.5%, and −29.7 to −57.4% at 7, 28, and 90 days, respectively. A noticeable trend was observed, indicating a decrease in concrete strength with a higher rate of aggregate substitution. However, all CX-E mixes achieved mean compressive strength results ranging between 43 and 61 MPa after 90 days. Among the C100/Y-E mixes, only C100/10-E and C100/20-E exhibited strength values exceeding 43 MPa at the same age.

The role of aggregate and water substitution on concrete strength in the current experiment may be also commented on. It is widely acknowledged in the existing literature that the use of recycled aggregates does not enhance this property due to the formation of a weak and complex interfacial transition zone, as well as its high water absorption, which compromises cement hydration and mechanical performance (Xiao et al. 2012; Agrela et al. 2013). However, Jiang et al. (2019) argue that fine particles can have a positive effect on replacements up to 50%, depending on the composition of the aggregates. In fact, by comparing the results of C100-E and C100/Y-E at 90 days, where the only difference lies in the use of recycled sand, it becomes apparent that fine aggregate replacements up to 20% had a positive effect on compressive strength.

Regarding the effect of treated wastewater on concrete strength, there is a divergence of opinions among researchers, enhancing the need for further investigation. Some authors argue that the presence of residual organic compounds has detrimental effects on compressive strength since it could have an effect over cement hydration and as a consequence on setting time and strength (Asadollahfardi et al. 2016; Ahmadi et al. 2017; Asadollahfardi & Mahdavi 2018; Catanzaro et al. 2019; Hassani et al. 2020a; Yahyaei et al. 2020). On the other hand, there are those who defend that the presence of residual organic solids and microorganisms can fill the voids and create effective interfacial interaction, thereby enhancing its performance (Swami et al. 2015; Oliveira et al. 2016).

The results depicted in Figure 5 demonstrate a decrease in the average strength of concrete as the level of aggregate substitution increases, aligning with findings from previous studies. All samples (except the control) had the effect of wastewater addition, which means that it is not immediately clear if it had a damaging or beneficial effect. Then, in further research, it might be necessary to investigate the individual and the interaction effect of treated wastewater and recycled aggregate aiming to clarify if the decline caused in compressive strength is associated with the use of only one or with both alternative materials.

Either way, the total effect (individual effects plus their possible interaction) of both parameters is compared with the control group in the current experiments. It is noteworthy to comment that the above-mentioned resistance results are compatible with several concrete applications, although it is essential to ensure its appropriate durability to guarantee safe utilization. This strength behavior is like those observed in several studies that used both wastewater and aggregate substitution: a general trend in mean concrete strength reduction (Elchalakani & Elgaali 2012; Ramírez-Tenjhay et al. 2016; Elchalakani et al. 2017; Ahmed et al. 2021; Raza et al. 2021). The mean strength decrease, evidently, varies between those studies, as many aspects (aggregate amount, size and type, wastewater quality, concrete mix design, cement type, and others) might have an influence over concrete's mechanical performance.

Nevertheless, it is important to note that the presented compressive strength results for each mix represent an average derived from five samples, indicating a range of outcomes. Consequently, in order to discern whether the discrepancies observed in the compressive strength results stem from natural variability or significant differences attributed to the various replacements tested, an ANOVA and post hoc Dunnet's test were conducted.

The analysis compared the effect of treated wastewater and coarse aggregate (CX-E versus C0) and treated wastewater and coarse and fine aggregate (C100/Y-E versus C0) replacements over compressive strength performance. Regarding the CX-E and C0 comparison, at 7 days, no statistically significant difference between the groups and the control sample was detected. As for 28-day results, coarse aggregate replacements of 20, 80, and 100% induced statistically significant differences, while at 90 days the same tendency was observed for substitutions exceeding 40%.

As for the C100/Y-E and C0 comparisons, the joint influence of drinking water, recycled coarse aggregate, and different ratios of natural fine aggregate were assessed. As anticipated, the presence of fine recycled aggregate and its higher water absorption led to a considerable decrease in compressive strength. After 28 and 90 days, all groups exhibited a statistically significant difference compared with C0, while at 7 days, only the groups C100/20-E and C100/30-E did not follow the same tendency.

In comparison with the control, the C100/Y-E mixes exhibited varying effects on splitting tensile strength at 28 days. Fine aggregate replacements ranging from 10 to 40% resulted in relatively lower reductions in strength (−4.0 to −16.2%), while a complete replacement of fine aggregate caused a significant decrease of −40.2%. Notably, no distinct trend was observed for higher proportions of fine recycled aggregate in this group. It is worth mentioning that C100/10-E and C100/40-E showed better results than C100-E. These findings align with the reports of previous studies (Gonzalez-Corominas & Etxeberria 2014; Bravo et al. 2015; Etxeberria & Vegas 2015; Etxeberria & Gonzalez-Corominas 2018; Plaza et al. 2021), which suggest that fine aggregate replacements up to 50% may lead to either increases or slight reductions in splitting tensile strength. This behavior is likely attributed to pozzolanic reactions, and the internal curing effect facilitated by ceramic waste, which contribute to a more efficient transition zone (Gonzalez-Corominas & Etxeberria 2014; Etxeberria & Vegas 2015). However, higher proportions of very fine recycled aggregate can compromise the cohesion between cement paste and aggregates, necessitate more water to maintain workability, and result in a more porous and less resistant concrete (Bravo et al. 2015) as observed in the C100/100-E mix.

Notwithstanding, the splitting tensile results for each mix are an average of five samples, so there is a range of data. ANOVA and Dunnett's test results comparing CX-E and C100/Y-E families with control C0 showed that, at 7 days, neither mix was statistically different from C0, while at 28 days only C80-E, C100/20-E, and C100/100-E were.

The results clearly demonstrate that the detrimental effects on water absorption and air-void ratio were more pronounced with higher proportions of recycled aggregates, with the C100/100-E mix exhibiting the poorest performance. This outcome aligns with the understanding that the physical properties of concrete are closely linked to the physical properties of the aggregates used. However, the influence of water quality on the air-void ratio and water absorption remains a topic of debate.

Some studies such as Ghrair & Al-Mashaqbeh (2016), Asadollahfardi & Mahdavi (2018), Brandão et al. (2019) and Gulamussen et al. (2021) reported lower air voids and water absorption in concretes incorporating treated wastewater. However, it is widely believed that impurities in wastewater can absorb water, subsequently affecting the formation of crystals and gels and resulting in a porous structure. This viewpoint is supported by studies conducted by Hassani et al. (2020a, 2020b), Raza et al. (2020, 2021), and Seyyedalipour et al. (2015).

Therefore, it is crucial to ascertain whether the observed decreases in water absorption and air-void ratio are from the combined effect of using alternative materials, as suggested by Elchalakani & Elgaali (2012), or if the effects of wastewater are variable and not substantial, with the reductions primarily associated with the utilization of recycled aggregates, as proposed by Raza et al. (2021). Further investigation is needed to shed light on this matter.

When comparing the CX-E and C100/Y-E families with the C0 reference, it is evident that the combined use of treated wastewater and mixed recycled aggregates, both coarse and fine, had a detrimental effect on the density property. The observed decreases ranged from −1.3 to −12.5%, with the highest reductions observed in the case of complete replacement of natural materials. Similarly, the saturated density showed variations in the range of −1.0 to −8.4%, while the true density exhibited variations from −0.7 to −5.7%. Despite those results, depending on the concrete's usage, the density reduction might not be considered a disadvantage. The ANOVA and Dunnett's test results indicate that only the C20-E mix did not show statistically significant differences in dry, saturated, and true density when compared with the C0 reference.

This research is significant because it addresses a gap in the existing literature about the use of both recycled aggregates and treated wastewater in concrete manufacturing. In summary, concrete with acceptable physical and mechanical properties was produced while using recycled aggregates from CDW and treated wastewater from a municipal sewage treatment plant as a source of water. As expected, the characterization of MRA revealed inferior properties compared with natural aggregates, while the quality of treated wastewater aligns with standard requirements for concrete manufacturing.

Our analysis also revealed that the use of treated wastewater caused a slight effect over the initial and final setting times. Concrete's consistency was negatively impacted by the replacement of natural aggregates, although only the C100/100-E mix displayed a significantly dry consistency, leading to difficulties during the casting process. Reflecting on the data presented, it becomes clear that the replacement of natural materials with recycled ones had a detrimental effect on concrete's mechanical properties. Compressive strength decreased with the substitution of drinking water and increasing amounts of coarse MRA and after 90 days, only the C20-E and C40-E mixes did not show statistically significant differences compared with the control (C0). In addition, at 90 days of age, the inclusion of small amounts of fine recycled aggregate (up to 20%) had a positive effect on this property. Regardless, the mean compressive strength results ranged between 26 and 63 MPa, and except for C100/100-E, all samples achieved compressive strength over 35 MPa, which enables the use of this type of concrete for several applications.

The mean splitting tensile strength at 7 days for the CX-E and C100/Y-E families exceeded that of the control sample. However, after 28 days, all samples exhibited lower results, with only the C80-E, C100/20-E, and C100/100-E mixes showing statistically significant differences. As for physical properties: water absorption, air-void ratio, and density experienced poorer outcomes when natural aggregates and drinking water were replaced with alternative materials.

The importance of this study lies in its potential to pave the way for a more conscious civil construction industry, as evidenced by the positive results achieved and the potential for conserving natural resources for more essential purposes. To further promote the use of concrete made with alternative materials, future research should focus on evaluating the durability properties and concrete-steel interaction to ensure the safety of using recycled materials concrete for structural purposes. In addition, it is important to understand the opinions of both the industry and consumers regarding this practice. Finally, it is important to highlight that all tests were conducted for research purposes and activities with different purposes should comply with current regional laws and standards.

This work was supported by the Coordination for the Improvement of Higher Education Personnel (CAPES) (Grant number 88887.513026/2020-00) and Brazilian National Council for Scientific and Technological (CNPq) (Grant numbers 132145/2021-9 and 311275/2015-0) and FAPESP (São Paulo Research Foundation, Grant number 2017/07490-4). The authors also thank SANASA and SBR Reciclagem for donating the treated wastewater and the recycled aggregate, respectively.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.