Characterization of residual sand removed from the grit chambers of a wastewater treatment plant and its use as fine aggregate in the preparation of non-structural concrete

The purpose of preliminary treatment in sewage treatment is to eliminate or minimize the adverse effects of debris on the functioning of the downstream processes and parts of the facilities. The sand removed from the grit chambers, together with the other debris, typically has between 35 and 80% solids in the dry stage, and 1 and 55% solids of volatile solids (WEF 1998). The average density of this material is of approximately 2.7 × 10⁶(100 mL)⁻¹E. coli and 0.6 helminth eggs g⁻¹ (Yamane 2007).

Although the presence of these residues impose difficulties in the management of WWTPs, with regard to the handling, treatment and final disposal, preliminary treatment has not received much attention, a fact that is reflected in the scarce scientific literature related to the subject and in the slow technological advancement proposed by manufacturers and designers.

In Brazil, the common practice regarding the disposal of sand from WWTPs is to discard it at the station's land area (burying or stacking) or send it to landfills. Overall, there is no concern related to the potentials associated with its direct use, reduction or recovery of the volume generated and related to environmental impacts. The disposal in Brazilian landfills, besides representing a cost of approximately US$ 70 t⁻¹ (value in May 2,014 accounting expenses associated with transport and disposal), occupies a volume that could be effectively intended for domestic waste.

By comparison, in the United States in some large WWTPs, the residue from grit chambers is incinerated together with other solids, such as grated materials and varied biological sludge. Some American states have environmental laws requiring lime stabilization prior to the disposal in landfills (Metcalf & Eddy 2003).

The potential sand reuse in the raw state that is without washing, drying and classification, is virtually non-existent. However, its use in civil construction is an alternative, provided that sanitation is performed to eliminate or significantly reduce the density of pathogenic micro-organisms and the removal of organic matter. This, however, is still a largely unexplored but promising subject, because besides the potential economic benefit, it implies in environmental gains in light of the possible reduction that could be achieved in the commercial sand extraction activity (for instance used in construction) of river beds/banks or other locations, which jeopardizes water quality.

In literature, there are few references regarding the use of sand removed from the grit chambers of WWTPs in Civil Construction. However, there are recent studies of another type of waste generated in WWTPs, as for instance the work developed by Chang et al. (2010), which evaluated the use of sewage sludge in the form of ash, as mortar and concrete element used in construction, together with commercial sand and cement. In that work, the researchers concluded that the addition of about 10% of this type of ash to the concrete mix reduces the amount of cement needed, without losing strength and consistency, thereby a good alternative for the final disposal of this type of waste.

In this context, our work evaluates the potential use of sand removed from wastewater treatment plants as fine aggregate in the composition of concrete for construction.

The research was conducted using sand removed from the primary treatment at the Monjolinho Wastewater Treatment Plant (WWTP), located at the city of São Carlos – Brazil (7.560.900 N, 198.000 E coordinates, 735 m altitude and average annual temperature of 20 ^oC). The WWTP has the capacity to treat an average flow of up to 636 L s⁻¹and maximum flow of 1.050 L s⁻¹, with the preliminary treatment consisting of

• mechanical rack grids with 2 cm spacing between bars;

• mechanical ladder grids with free spacing between 3 mm blades;

• 3′ Parshall flume;

• aerated grit chambers with simultaneous O&G (oil and grease) removal;

• aerobic system for degrading O&G removed in the aerated grit chambers;

• gas treatment.

The settleable residue is removed at regular intervals from the grit chambers by submersible pumps and routed to sand ‘classifiers’ equipment, and then to the dump-cart.

According to operational data from January 2012 to March 2014, the WWTP had an average flow of 527 L s⁻¹, resulting in the following parameters for each grit chamber: air flow rate of 5 m³ min⁻¹, mean application rate of 720 m³ m⁻² day⁻¹ and hydraulic retention time of 6 minutes.

The amount of sand removed during the study was of 25 t month⁻¹, corresponding to 18.3 kg (1,000 m³)⁻¹. The current per ton price for transporting and disposing of this waste in the landfill of São Carlos is of about US$ 70 (transport US$ 29 and disposal US$ 41).

The characterization of residual sand was based on the analyses for homogenized samples (six, at different times of the year), of about 70 kg. The samples were collected immediately after the waste was placed in the dump-cart and sent to the experimental agricultural greenhouse unit (6.0 m long, 4.0 m wide and 3.0 m high).The exhaled gases were removed by an axial exhaust fan, 1,500 m³ h⁻¹ capacity and service pressure of 6 mmca – after which determinations of total solids, moisture and total and fecal coliforms were performed.

For determining the total and fecal coliforms, 20 g of residual sand with 200 mL of deionized water were mixed for 5 minutes. After mixing, the sample was kept at rest for 24 hours, after which the supernatant was analyzed according to the procedure in ‘Standard Methods for the Examination of Water and Wastewater’ (APHA 2005).

Determination of particle size distribution was also performed for two of these samples. Next, the concentration of volatile and fixed total solids was determined in order to verify the percentages of organic matter and inert residue of the fractions retained in this test.

Owing to the high organic material content, the use of raw sand (without washing, drying and classification) to be applied in construction is not recommended. Thus, the sand cleaning and drying procedure was performed in the experimental unit, in order to reduce the moisture, organic matter and the pathogenic micro-organism density, making it safe from a microbiological point of view for the desired potential applications.

The residual sand was sieved (1.18 mm sieve aperture) to remove the bulkier materials. Next, washing with clean water and sodium hypochlorite at 12% concentration was performed in a mixer in order to inactivate the microorganisms and the possible oxidation of organic matter in the sand aggregate. Finally, the washed sand was placed in four draining units (each one 0.6 m long, 0.6 m wide and 0.2 m in effective height) for drying. Figure 1 shows the flowchart of the general procedure used in the research.

After the cleaning and drying procedure, the following determinations were carried out on four samples: grain size distribution, mass density, apparent density, number of total solids, moisture and total and fecal coliforms. These determinations were also made for the commercial reference sand to compare the results. The reference sand also underwent the sieving step (1.18 mm aperture) before the analyses were performed.

In order to evaluate the potential use in constructions, cylindrical test samples (100 mm in diameter and 200 mm in height), with 1:3:2 trace (cement:sand:gravel) were prepared to evaluate the mechanical strength. Before the specimens were molded, the Slump Test was performed to determine the consistency of the truncated cone. After demolding, the specimens were sent to a humid chamber at room temperature (23 ± 2) °C, and air relative humidity higher than 95%, where they remained during the curing time (28 days).

The concrete components used were: Pozzolanic Composite Portland cement with characteristic strength of 32 MPa (CP II–Z–32), gravel No. 1 (grain size ranging from 9.5 to 19 mm), residual and commercial sand and water (water/cement ratio was set at 0.5–0.7).

Test specimens were made with concretes prepared by the total replacement of the common sand for the residual and partial sand at the proportions of 20, 30, 50 and 70%, which were subjected to the axial compressive strength and tensile strength by diametral compressive tests at 28 days of age.

To determine the characteristics of the concrete aggregates and the molded specimens, the following Brazilian test methods and its international equivalent, but not necessarily equal, were used:

• Unit mass: ABNT NBR NM 45:2006 (ASTM C29/C29M:2009);

• Apparent unit mass: ABNT NBR NM 52:2009 (ASTM C 128:2012);

• Grain size: ABNT NBR NM 248:2003 (ASTM C33/C33M:2013);

• Slump Test: ABNT NBR NM67:1998 (ASTM D6005:2009);

• Molding and curing of sample specimens: ABNT NBR 5738:2003 (ASTM C192/C192M:2013)

• Axial compressive strength: ABNT NBR 5739:2007 (ASTM C39/C39M:2014);

• Compressive and diametral tensile strength tests: ABNT NBR 7222:2011 (ASTM C496/C496M:2004)

Table 1 shows the determination results related to residual sand removed from the grit chambers, before and after the cleaning and drying procedure, and of the reference sand.

According to the data presented in Table 1, it was found that

• Before the cleaning and drying procedure, the residues showed high percentage of total fixed solids, high moisture content and expressive density of total and fecal coliforms, demonstrating the need for treatment in order to eliminate or significantly reduce pathogenic bacteria density, to make it safe from a microbiological aspect, as well as to decrease the amount of water and organic material, thereby enabling the desired potential applications.

• The cleaning and drying procedure used was effective, achieving the following removal efficiencies: about 98.8% of moisture, 67.1% of total volatile solids and 4-log E. coli.

The residual sand after the cleaning and drying procedure showed similar characteristics to those of the reference sand, showing only a higher percentage of total volatile solids.

Figure 2 shows the percentages of total volatile and fixed solids of the residual sand retained in sieves in the grain size test before the cleaning operations¹.

The fractions retained in the smaller aperture sieves exhibited lower organic waste concentrations, showing the importance of the preliminary screening step to reduce the amount of organic matter from the residual sand, which used in this study 1.18 mm apertures.

The characterization results of the fine and coarse aggregates used in the concrete composition were

• Residual sand: unit mass 1.45 kg L⁻¹, specific mass density 2.44 g cm⁻³, maximum diameter 1.18 mm and fineness modulus 1.57;

• Reference sand: unit mass 1.60 kg L⁻¹, specific mass density 2.62 g cm⁻³, maximum diameter 1.18 mm and fineness modulus 1.67;

• No. 1 gravel: unit mass 1,57 kg L⁻¹, maximum diameter 6.30 mm and fineness modulus 4.84.

These results showed that the residual sand and reference sand had very similar characteristics. Both sands were classified as extremely fine (fineness modulus <2.4). Note that a high percentage of fine material requires increased mixing water and consequently of cement, for the same water/cement ratio, thereby making the concrete more expensive. In addition to this, material less than 0.076 mm can be mixed with the cement, producing discontinuity in the mixture and reducing the concrete strength. On the other hand, concrete without the fine material has low consistency, consequently subject to increased permeability and aggressive agents.

Table 2 shows the results of the axial compressive and tensile strength by diametral compressive test performed on the sample specimens that were prepared by substituting the commercial sand for the residual sand (as fine aggregate) at the following proportions: 100, 80, 70, 50, 30 and 0%.

According to the results shown in Table 2, it was found that

• the test specimens prepared with reference sand showed much higher strength values (f_ck) characteristic to compressive and tensile (f_ctk) concrete, than those prepared with 100% residual sand;

• there was an increase in the values of f_ck and f_ctk with the increasing amount of commercial sand. For the characteristic compressive strength, increases in the tests with 100% residual sand were of 46.6%, 63.7%, 86.9%, 127.1% and 165.1%, respectively, for the following percentages of commercial sand added: 20, 30, 50, 70 and 100%. For the f_ctk values, for the same proportions of common sand added, the increases were of 69.3%, 89.0%, 93.6%, 109.9% and 97.0%, respectively;

• the tensile strength of the concrete varied between 8.2% (100% commercial sand) and 12.8% (substitution of 20 and 30% of residual sand for commercial sand) of compressive strength;

• the main application for the concrete made with the sand extracted from the WWTP is the construction of non-structural elements, e.g. sidewalks, curbs and flooring, since its samples presented similar physical characteristics to the commercial sand and the strength results were satisfactory, considering that the f_ck values required in Brazil for simple concrete made from recycled residues from construction activity are within the range of 10–15 MPa.

Given the above, it appears that the concrete prepared with 80% mixture of residual sand and 20% commercial sand as fine aggregate, showed an acceptable f_ck value (10.8 MPa), considering non-structural applications. However for safety reasons, in order to increase concrete durability, the proportion of 70% residual sand and 30% commercial sand is recommended, since the characteristic compressive strength of concrete was 12.0 MPa.

This work verified the feasibility of using the residual sand removed from grit chambers in WWTPs as fine aggregates in the preparation of non-structural concrete, provided washing and drying is done, in order to eliminate or significantly reduce the density of pathogenic micro-organisms and remove organic matter.

The characterization of the settleable residue showed that to use this type of material requires an adequate treatment, in order to reduce the density of pathogenic micro-organisms, moisture and organic matter, given that the residue showed high moisture content (14.8%), significant amount of total volatile solids (3.1%) and expressive total coliform (3.84 × 10⁷) and fecal (5.22 × 10⁵) densities.

The cleaning and drying procedure of the residual sand used in the study was effective as it achieved the following removal efficiencies: about 98.8% of moisture, 67.1% of total volatile solids and 4-log of E. coli. The following steps were used in the procedure: sieving, washing (clean water and sodium hypochlorite) and drying the residual sand.

To increase the durability of concrete, for safety reasons, it was proved that due to the higher strength values, the partial replacement of residual sand for ordinary sand was an excellent alternative. And using 70% of residual sand as fine aggregate was proved feasible, since the strength value was adequate, in addition to preventing the disposal of much of this waste in sanitary landfills.

The authors are grateful to: FAPESP – Fundação de Amparo à Pesquisa do Estado de São Paulo, for financial and research support; to the technical staff of the Monjolinho WWTP of São Carlos for their assistance throughout the research; and tothe technical staff at the Sanitation and Civil Construction laboratories of USP São Carlos for their willingness, encouragement and assistance during the analyzes and tests.