Biosand reactors as a feasible solution for the treatment of winery wastewater (Part 1)

Abstract

Based on over 10 years of laboratory and pilot scale studies, a prototype biosand reactor system for remediation of winery wastewater is proposed. In these systems, organics are bio-degraded by microbial communities and acidic winery wastewater is neutralised via calcite dissolution into calcium and carbonate. The calcium released into the effluent also decreases the sodium adsorption ratio, protecting the receiving environment from sodicity.

By using locally available sand with particles ≤0.425 mm removed, systems can be designed to adjust the flow rates to achieve the desired hydraulic retention time for optimal remediation of winery wastewater.

This article reports a basic synopsis of the practical findings of the research. In-depth scientific results are available either online or by request from the corresponding author.

Introduction

Winery wastewater (WWW) is classified and regulated by the Department of Water and Sanitation (DWS) as “biodegradable industrial wastewater”. In many wineries in South Africa, WWW is used for irrigation after the solids have been reduced in settling deltas or oxidation ditches and the pH has been adjusted. Depending on the effluent’s quality and quantity, this can negatively impact the receiving environment over time. Passive treatment systems such as constructed wetlands (CWs) have been used to remediate WWW. However, plants can die when exposed to high concentrations of polyphenolics, which are often present in WWW. Constructed wetlands typically contain gravel, sand, or both as the media for the attachment of the functional microbial biofilm and plant growth. Biosand reactors (BSR) are essentially unplanted CWs that are easier to maintain than their planted counterparts.

A number of laboratory-based and pilot studies have been conducted to determine the effectiveness of BSRs for the remediation of WWW. Lessons learned from these studies have led to the development of a prototype. Results of the most recent studies and their practical applications are briefly presented in this report.

Set-up and operation of pilot biosand reactors

Biosand reactor systems consist of infrastructure for primary settling, WWW equalisation, and biological and physicochemical treatment. The latter two functions occur in one or more BSR modules, as shown schematically in Figure 1. Modules can be added to the system to scale up treatment capacity.

Good flow rates can be achieved when applying the WWW to the top of the sand and allowing it to percolate through the sand, in other words, when the BSR modules are operated vertically. In the sand matrix, (1) bioremediation (removal) of the organics takes place, and (2) acidic WWW is neutralised without having to add chemicals. A unique design feature is the adjustable outlet, which controls the flow rates by increasing or decreasing the hydraulic head within the BSR modules. The higher the hydraulic head, the higher the water pressure in the sand and the higher the flow rate. An example of how this is applied is during the crushing season when WWW treatment systems need to deal with high organic loads. More organics equate to more substrate (food) for the microbial populations. The microbial biofilm, therefore, proliferates and reduces the pore spaces in the sand, decreasing the flow rate. This can be counteracted by increasing the water pressure by temporarily manipulating the height of the outlet until the organic load and biomass decrease after the crush period.

FIGURE 1. Cross section of a biosand reactor module (adapted).¹

In wineries where sufficient solids are removed from the existing primary settling infrastructure, the WWW can be directly applied to the BSR surface. If the WWW has high concentrations of solid particles, secondary settling may be necessary. The set-up of a pilot system requiring infrastructure for additional settling is depicted in Figure 2. Apart from the solids settling function, using additional settling tanks has an advantageous buffering effect because it reduces the variability of the influent to the BSRs.

FIGURE 2. Layout of a pilot biological sand reactor winery wastewater treatment system showing the flow of winery wastewater from settling basin.¹

Functionality of biosand reactors

Removal of organics and neutralisation of acidic winery wastewater

The main functions of BSRs are to reduce the chemical oxygen demand (COD) concentrations and increase the pH of acidic WWW. These parameters were measured in a pilot system over two crush seasons, shown in Figure 3.¹ During the start-up period, which coincided with the first crush period, the functional microbial communities were slowly acclimatising to the BSR environment and the WWW. Once established, consistently high COD removal results were obtained during both crush periods (Figure 3a). The pH of WWW is highly variable, but the pH of acidic WWW is increased to near-neutral values in BSRs (Figure 3b). It was found that the major alkalinisation mechanism is via the dissolution of calcite a (mineral form of calcium carbonate) and, to some extent, via microbial processes.² Another valuable effect of calcite dissolution is the addition of calcium to the WWW, which decreases the sodium adsorption ratio (SAR, Equation 1). This is important when the WWW contains sodium. Addition of the divalent cation Ca²⁺ mitigates against sodicity caused by excess monovalent Na⁺ cations.

Based on an average influent pH and results from extensive column and laboratory-based experiments, it was calculated that the calcite in the sand of BSRs treating WWW with a pH of 3 would be expended after 82 years, well beyond the projected ‘life’ of the sand. In reality, it is recommended that the sand is replaced every 10 to 15 years.

⁽¹⁾

FIGURE 3. Chemical oxygen demand (a) and pH (b) measurements from a pilot biosand reactor system during two consecutive crush periods.¹

Flow rates

The flow of WWW through BSRs reduces after start-up because the functional microbial biofilm reduces the pore spaces between the sand particles. Microbial growth is a positive feature but requires consideration because of the knock-on effect on flow rates in BSRs. In an earlier pilot system that was operated in horizontal mode, the flow rates were retarded to the extent that the volume of WWW that could be treated was lower than design levels. Indeed, many stakeholders have queried whether BSRs become clogged with solids and whether they need backwashing. Studies have shown this is not the case and that the reduction in flow rate is positively correlated with COD loading and is naturally reversible after the crush period and/or lower loading as the organic solids degrade. Slow flow rates can be overcome by employing vertical operation, increasing the hydraulic head and regularly scarifying the surface of the sand.¹ More recently, it was found in column experiments that if smaller sand particles (≤0.425 mm) are removed, the hydraulic conductivity of the sand can be increased a further 10-fold without having a negative impact on COD removal performance (Figure 4).³

Indeed, the achievable flow rates in systems containing fractionated sand are too high for effective WWW remediation if non-controlled gravity feeding is employed. This is because a minimum contact time (hydraulic residence time) is needed between the incoming WWW and the microbial biofilm attached to the sand. In the proposed prototype, an optimal hydraulic residence time will be achieved by adjusting the BSR outlet to increase or decrease the flow according to need, as shown in Figure 1.

FIGURE 4. Comparison of hydraulic conductivity (a) and COD removal (b) of raw sand and fractionated sand (particles ≤0.425 mm removed).³

Biosand reactors for wineries of different sizes

Based on our studies’ results and the projected WWW volume generated for wineries crushing from 10 to 5 000 tonnes of grapes per annum, the number of BSR modules with an internal sand capacity of 5.6 m³ required to treat WWW was calculated (Table 1).³ The calculations accounted for two different peak flows: wineries that generate 80% or 50% of their effluent during the crush period. The data suggests that using the modular set-up, the systems are attractive for small to medium-sized wineries.

The advantages of these systems over conventional systems are that:

• after the initial start-up period, the systems are resilient to lengthy periods of inactivity without losing functionality;
• they require minimal maintenance;
• skilled operation is not required; and
• fiscal requirements for operation are minimal because they are gravity-fed and not aerated.

TABLE 1. Treatment capacity and number of biosand reactor modules required for different wineries (adapted from Holtman et al., 2023).

Funding

This work was supported by the Wine Industry Network of Expertise and Technology (Winetech) (CSUR 13091742538). Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and the funding entities do not accept any liability in this regard.

References

1. Holtman, G.A., Haldenwang, R., Welz, P.J. (2022a). Comparison of continuous and pulse mode of operation of pilot biosand reactors treating winery effluent. Ecological Engineering. 182: 106706 https://doi.org/10.1016/j.ecoleng.2022.106706
2. Holtman, G.A., Haldenwang, R., Welz, P.J, (2022b). Calcite Dissolution and Bioneutralization of Acidic Wastewater in Biosand Reactors. Water 14, 3482. https://doi.org/10.3390/w14213482
3. Holtman, G.A., Haldenwang, R., Welz, P.J. (2023). Biosand reactors for remediation of winery effluent in support of a circular economy and the positive effect of sand fractionation on hydraulic and operational performance. Journal of Water Process Engineering 53: 103849 https://doi.org/10.1016/j.jwpe.2023.103849
4. Holtman, G.A., Haldenwang, R., Welz, P.J. (2022). Effect of particle character and calcite dissolution on the hydraulic conductivity and longevity of biosand filters treating winery and other acidic effluents. Water 14: 2603 https://doi.org/10.3390/w14172603

For more information, contact Gareth Holtman at gareth@holtman.co.za.

Click here to get your copy of WineLand Magazine and here to subscribe to our newsletter.