A bioflocculant-supported dissolved air flotation system for the removal of suspended solids, lipids and protein matter from poultry slaughterhouse wastewater

Dissolved air flotation (DAF) is a wastewater treatment technique employed in the separation of low-density solids from wastewater (Al-Shamrani et al. 2002). This technique is highly dependent on the suitability of the sparging system used in the DAF. If sufficient pressure is applied to the diffusers, microscopic bubbles are formed when the wastewater is pumped into the flotation cell. The micro-bubbles generated in the system can be harnessed for the removal of fats, oil and grease (FOG), suspended solids and biomass (Amaral Filho et al. 2016). The good effectiveness of this system can be attained when the size of the bubble is suitable when compared with the size of the particles that have to be separated; furthermore, compatibility of the surface charges for both the suspended particles and the micro-bubbles can also influence the DAF's operational efficiency (Han et al. 2007). The quantity of the micro-bubbles generated, including their size, is dependent on the pressure applied to the air diffusers.

The addition of flocculants into the DAF can also enhance the efficiency of the system. In most cases, flocculation–flotation methods involve the addition of chemical flocculants into the system. Chemical flocculants such as aluminium sulfate and ferric chloride, as well as polyacrylamide, have been determined to be highly efficient. However, they are expensive, non-biodegradable, and can have a harmful effect on the environment; moreover, they have been determined to be toxic to humans (Wang et al. 2010). These considerations prompted researchers to assess alternative solutions with minimal impact on the environment. Hence, bioflocculants have received much attention lately. These polymers are biodegradable, environmentally friendly and they do not pose a risk to human health. This part of the study mainly focuses on the application of bioflocculants produced by isolates from the poultry slaughterhouse wastewater (PSW), for their use in DAFs for the removal of suspended solids, FOG and protein matter from the PSW. The aim of the study was to design a bench-scale bioflocculant-supported DAF (Bio-DAF) and assess its efficiency in removing total suspended solids (TSS), protein and FOG from PSW; and to compare the efficiency of the Bio-DAF to those processes in which chemical flocculants are used, including a conventional DAF (control).

Comamonas aquatica (BF-3) and Bacillus sp. BF-2 were previously isolated as reported in Dlangamandla et al. (2016). These isolates were determined to produce bioflocculant with a high flocculation activity. As such, they were used to in the Bio-DAF systems designed.

A DAF system was designed (see Figure 1). The DAF tank was designed to have a 2 L capacity. A Resun air pump (Ac 9906) was used to pump air into the DAF tank using silicone tubing. The silicone tubing that pumped air into the DAF tank was connected to four air diffusers to provide sufficient micro-bubble formation in the system. The air was pumped at 16,000 Pa. PSW was pumped from the feed tank into the system using a Gilson^® mini plus evolution pump at 0.11 mL/min and at a hydraulic retention time of 12 days. The system contained skimmers to skim FOG, and proteins including suspended solids floating on the surface of the wastewater. The suspended solids skimmed off were washed down the DAF tank's side by gravity, using four openings in the system that collected spillages from the overflow. The effluent was pumped out of the tank using a Watson-Marlow 502 peristaltic pump at the same rate as the influent to a sampling container. Dimensions of the Bio-DAF used are listed in Table 1 – see Figure 1 for a schematic illustration of the experimental set-up.

A volume (50 L) of the PSW was collected from a poultry slaughterhouse once every 2 weeks, using sterile 25 L polypropylene bottles. The wastewater was analysed for pH, temperature, total dissolved solids (TDS), total suspended solids (TSS), turbidity, FOG, protein concentration, total chemical oxygen demand (tCOD) and soluble chemical oxygen demand (sCOD) concentration. Prior to the addition of the wastewater to the DAFs, bioflocculants were produced, screened and characterised. For every 2 L of the collected wastewater, 20% (v/v) of the culture of bioflocculants produced by the isolates C. aquatica and Bacillus sp. BF-2 grown over 48 h was added to the PSW after which the air diffusers were switched on. The micro-bubbles created by the high pressure applied by sparging into the system elevated the solids to the PSW surface, where they were skimmed off by stainless-steel skimmers agitated at 66 rev/min using a Dragon Laboratory OS20-S agitator (Gulas et al. 1978; Zabel 1985). The samples were collected periodically at 48 h intervals from the DAF system and analysed immediately.

The TSS were measured using the ESS method 340.2 (ESS 1993). The TDS were analysed with the multi-parameter PSCTestr 35 (Wirsam Scientific, Malaysia). The turbidity of the samples was quantified by the Turbidimeter TN-100 (Wirsam Scientific, Indonesia). The COD test is an indirect measure of organic compounds, which operates on the basis that the strong oxidising agent (potassium dichromate) fully oxidises the organic matter in the presence of a 50% strong acid solution (sulfuric acid). The COD was quantified using Merck COD solutions: A (cat. no. 1.14679.0495) and B (cat. no. 1.14680.0495). A Merck Spectroquant^® NOVA 60 was used to measure the COD concentrations. FOG were analysed at the City of Cape Town's Scientific Services Laboratory according to the American Public Health Association standards. Furthermore, the primary basis for the Bradford's assay is to detect the presence and concentration of proteins in a solution based on the absorbance shift, using a dye which changes colour from reddish to blue when proteins are present in a solution and the protein concentration is observed at a wavelength of 595 nm in a spectrophotometer. Overall, protein concentration was quantified using the Bradford's assay.

In order to achieve the aims of the study, DAFs were designed to minimise clogging when the wastewater is fed to the treatment system and to reduce diffuser fouling. The DAF system was operated using bioflocculants and a chemical flocculant, and conventional DAFs were used for control studies, that is, without flocculant supplementation, to observe which system had a competitive advantage. Three states were observed, namely, unsteady state (from day 0 to 2), transition to steady state (day 3), and steady state (day 4 to 10). Therefore, the DAF efficiency was quantified during the steady state.

A TSS removal method is a familiar technique utilised to determine particulates that cannot be filtered using conventional filters. These solids are usually bigger than 2 μm, whereas for TDS, smaller particulate matter less than 2 μm is quantified. TSS are made up of organic and inorganic matter present in the wastewater. This parameter is important when observing water quality and it is indirectly correlated to turbidity. When a high quantity of suspended solids is present in the wastewater, the water becomes turbid. By description, turbidity is the determination of light scattering in the water sample (Hannouche et al. 2011); hence, the turbidity of the PSW can be directly affected by both suspended and dissolved solids.

Figure 2(a) illustrates the removal efficiency of TSS by Bio-DAF and chemical-supported DAF (Chem-DAF) at steady. When observing Bio-DAF at unsteady steady (Figure 3(a)), it removed 7% and 56.5% TSS on day 0 and day 2, respectively. When observing the steady state, the average TSS removal for the Bio-DAF was 91%. This was attributed to the microbial isolates introduced into the system – an indication of their competitiveness compared with other microorganisms present in the PSW. A 48 h acclimatisation period for the isolates was required for the Bio-DAF system. Such an operational strategy can be ensured when the isolates are cultured for 48 h prior to the initiation of the PSW inflow, thus enabling the production of active bioflocculants. Overall, the isolates selected and used in the Bio-DAF were able to produce and sustain bioflocculant production, thus modifying surface charge of the suspended solids, which enabled macroflocs to be formed. This was followed by using stainless-steel skimmers (Dragon Laboratory OS20-S agitators) at 66 rev/min, to skim the flocculated solids. For the chemical flocculant-supported DAF (Chem-DAF), 2% (m/v) alum [KAl(SO₄)₂.12H₂O] was used, which was added to the PSW, resulting in the successful suspension of TSS in the wastewater, a phenomenon also attributed to the interaction of negatively charged suspended matter and the positively charged metal ions which neutralised the negatively charged suspended particles, forming floculatable flocs in the PSW. The buoyancy of the particles (assumed to have been encapsulated by a film of FOG) was hypothesised to have been assisted by FOG presence in the PSW, which resulted in the increased removal efficiency of the suspended solids so that the smaller flocs attached to larger flocs rising to the surface of the wastewater at a higher velocity. Owing to the immediate availability of surface charge in the Chem-DAF, in comparison with the Bio-DAF, which needed a 48 h acclimatisation, the Chem-DAF at unsteady state reduced 71% to 77% of TSS. During the unsteady state, the Chem-DAF was more efficient as compared with the Bio-DAF. Under steady state, the average TSS removal achieved by the Chem-DAF was 84% in comparison with the 91% achieved by the Bio-DAF. Overall, the average TSS removal achieved by the Chem-DAF used in this study was higher than the 34% removal of the TSS obtained by Amuda & Alade (2006), using alum as a flocculant in a jar test, where the DAF dose was 2 mg/L of alum – a concentration similar to that used in this study.

In control studies, a conventional DAF (Conv-DAF) system, without chemical or bioflocculant supplementation, also resulted in reduced TSS in the PSW. At the steady state, this system removed an average TSS of 33%, which is lower than the TSS removed by both a chemical and bioflocculant-supported DAF system (see Figure 2 – additional information). In both the Chem-DAF and Bio-DAF, TDS and turbidity reduction were also assessed. At steady state, the Bio-DAF achieved a TDS reduction of 76%, including turbidity reduction of 98%, while the Chem-DAF only obtained a reduction of 22% TDS and 97% turbidity, whereas the Conv-DAF only achieved a reduction of 26% and 97% TDS and turbidity, respectively – a result attributed to the direct influence of flocculant supplementation in the DAFs.

In the Bio-DAF, there was a conventional correlation between TSS, TDS and turbidity. As more TSS was removed, turbidity improved, while TDS was reduced. The low TDS removal in the Conv-DAF system was attributed to the absence of flocculants, with the micro-bubbles being unable to facilitate the attachment of TSS as a consequence of the presence of FOG. As the objective of this part of the study was to asses the removal of TSS in the PSW, the resultant effluent was determined to be in accordance with the South African by-law discharge standards for TSS (<1,000 mg/L). Overall, a higher TSS removal was achieved when compared with the 37% TSS removal efficiency achieved in a Conv-DAF (Del Nery et al. 2007). Similarly, in a study by De Nardi et al. (2011) on the efficiency of a DAF system supported by a cationic polymer as a flocculant for the treatment of PSW with similar characteristics as the PSW used in this study, only 65% TSS reduction was achieved, which is lower than the 91% TSS removal obtained using the Bio-DAF.

Lipids can be defined as oils, grease, fats and long chains of fatty acids. These lipids are the constituents of wastewater, mainly from the food-processing industry. Their presence in wastewater as organic material decreases dissolved oxygen (DO) of the wastewater and enhances levels of biochemical oxygen demand and COD (Chipasa & Mędrzycka 2006). In biological wastewater treatment plants, lipids are degraded by an enzyme called lipase (Andersson 1980). During the unsteady state, Figure 3(b), the removal of lipids in the Bio-DAF was between 91.3% and 91.6%, compared with 82% to 92% achieved by the Chem-DAF. During the steady state operation (Figure 2(b)) of the system, the average lipids' removal efficiency slightly increased to 93% for Bio-DAF when compared with the 92% obtained by both Chem-DAF and Conv-DAF systems. FOG contains a carboxyl group that is negatively charged; hence for flocculation to occur it requires an oppositely charged flocculant (Vance & Vance 2008). As the Bio-DAF contained bioflocculants from a co-culture, with different functional groups, the bioflocculants produced, particularly those produced by C. aquatica (protein constituents), had functional groups such as amino (positively charged) as well as carboxyl (negatively charged), while the other bioflocculants produced by Bacillus sp. BF-2 were negatively charged; hence, the FOG could attach to both the positively charged site of the C. aquatica bioflocculants and the negatively charged BF-2 flocculants, which could improve floc formation. Generally, Bacillus sp. produces lipases, which can break down FOG, which in turn influences the removal of lipids in the Bio-DAF; hence there was higher reduction of FOG in the Bio-DAF as compared with the Chem-DAF and Conv-DAF (Eggert et al. 2003). Similarly, the Chem-DAF contained a positively charged flocculant; hence, flocculation of the lipids was high although slightly lower than that of the Bio-DAF. It was previously determined that the Bio-DAF can remove a high quantity of lipids compared with the Conv-DAF, particularly when the separation of FOG is facilitated by the supplementation of flocculants into a DAF system, with the literature indicating that such supplementation can result in FOG removal of between 63% and 99%, and between 80% and 99% for fat (De Nardi et al. 2008). This study proved that the Bio-DAF could remove lipids slightly more efficiently than the Chem-DAF.

Proteins are known as the basic components of life: they are composed of nitrogen, hydrogen and oxygen. They are present in wastewater as soluble microbial products, and proteolytic enzymes produced by the microorganisms during their metabolic processes. The enzymes are produced by the microorganism extracellularly to degrade organic matter present in the wastewater so that the cells can utilise it as a nutrient source. They are major components of total organic carbon and total organic matter. High N:C ratio indicates the presence of proteins in the wastewater (Westgate 2009). The nitrogen and carbon concentrations of the PSW used in this study were 211 and 546 mg/L, respectively, which resembled PSW characteristics in studies carried out by Basitere et al. (2016) and Bustillo-Lecompte et al. (2016). These soluble proteins can be removed by coagulation and flocculation; however, in wastewater that contains FOG, a DAF system is also required. The proteins that are recovered by a flocculation–flotation system used to treat the wastewater cannot be reused, as the flocculants that attach to the proteins change their structure, which in turn changes their function (Bialas et al. 2014). In this study, soluble proteins were quantified to assess whether the system designed could effectively reduce total protein concentration prior to the wastewater's being biologically treated. The untreated influent contained 70 mg/L of soluble proteins. Figure 2(c) illustrates the protein concentration in the PSW prior to treatment using Bio-DAF, Chem-DAF and Conv-DAF. At unsteady state, Figure 3(c), the Bio-DAF removed 76% to 77% of proteins, with similar removal rates observed for the Chem-DAF, which removed 77% to 77.3%. An improvement to 79% (Bio-DAF) was achieved during the steady state, whereas 71% of proteins were removed by Chem-DAF, with the Conv-DAF removing 78% of soluble proteins. The Chem-DAF achieved the lowest protein removal compared with the Bio-DAF and Conv-DAF. Protein charge is dependent on the pH of the solution; it can carry a net positive or a net negative charge. In this part of the study, the Bio-DAF removed high protein concentration as compared with the Chem-DAF; this was because bioflocculants contain different functional groups to which the protein could be attached, resulting in efficient flocculation, whereas the Chem-DAF contained a flocculant that carried a single charge.

High COD concentration in wastewater means there is high quantity of oxidisable organic matter that leads to deterioration of DO in the wastewater, further creating an anaerobic environment which endangers aquatic life. sCOD determines the biodegradable part of COD, while tCOD determines the non-biodegradable portion in the wastewater. The Bio-DAF reduced 62% tCOD as well as 57% sCOD, whereas the Chem-DAF reduced only 52% tCOD including 53% sCOD. The Conv-DAF reduced 57% as well as 63% of tCOD and sCOD, respectively. The Bio-DAF was operated at high pressure and slower flow rate that in turn increased the residence time and mixing (Hami et al. 2007) – an operational strategy suited for effective COD reduction. Flocculation activity is affected by time and mixing; that is, when the residence time was increased, the microorganisms inoculated into the wastewater were exposed to more mixing; hence this system was effective for COD reduction. These data depicted that aerobic bacteria inoculated into the wastewater were capable of degrading these organic contaminants (Magnaye et al. 2009). In the Conv-DAF, these results depicted that there was high tCOD removal compared with sCOD removal in the system. Since this system was a chemical system, it achieved the highest removal of the non-biodegradable COD. In the Conv-DAF system, sCOD reduction was higher compared with that of the Bio-DAF. The tCOD reduction of the Conv-DAF and Chem-DAF was lower compared with the tCOD reduction achieved by the Bio-DAF – a phenomenon which needs further exploration.

The use of a DAF system has been described. This study focused on the design and efficiency of Bio-DAFs, compared with a Chem-DAF and Conv-DAF for TSS, FOG and protein removal. The results showed that the Bio-DAF removed 91% TSS, 93% FOG and 79% protein, while the Chem-DAF only removed 84% TSS and 92% FOG, including 71% protein. Therefore, it was proved that the Bio-DAF was more efficient than the Chem-DAF and Conv-DAF in the removal of TSS, FOG and protein present in the PSW. As a Bio-DAF is a fairly new design concept, it is recommended that its functionality, and thus performance reliability, must be assessed under varying shock loadings.