Optimization of coagulation-flocculation process in the treatment of wastewater from the brick-manufacturing industry

Several industries require water of varying degree of quality and quantity for effective and efficient operation. Sustainable supply of water of right quality, in the right quantity, at the right time and at the right place is needed for an effective industrial operation (Payne 2007). Manufacturing industries need reliable source(s) of water supply in order to function effectively. The manufacturing industries in the United Kingdom represent the largest non-domestic user of mains water. In England and Wales, the manufacturing industry uses 25% and 28% of mains water respectively, while in Scotland, the figure is as high as 41% (WRAP 2011). According to UN Water statistics, the manufacturing industry demand for water is estimated to increase by 400% by 2050 (UN Water 2014a). Hence, in addition to facing the challenge of waste minimization and pollution control, manufacturing industries are also faced with the increasing challenge of limited water supplies (Gumbo et al. 2003). Manufacturing industries that are heavily dependent on potable water supply from mains are at higher risk of exposure in the face of the increasing water scarcity. This is because public water supplies will tend to cut off supplies to industries rather than households as potable water scarcity heightens (Lambooy 2011). Water is essential for a wide variety of manufacturing operations; however, the pressure on water resources by industries is mainly compounded by the impacts of wastewater discharges and their pollution potential than by the quantity used in production (UNESCO 2009).

The brick making industry is one of the industries that uses water as a major raw material in its production process. Brick has been a reliable construction material for centuries and its excellent sustainability credentials and affordability have made it a common choice in the construction industry. Bricks are essentially produced by mixing water and ground clay, and forming the mixture into the desired shape (Bricks Industry Association 2006). In the United Kingdom (UK), the clay brick sector produced around 1,554 million bricks in 2012 from 60 brickworks across the country (Bricks Development Association (BDA) 2013). However, brick manufacturing is a water-intensive process; hence, water resource efficiency plays a major role in the sustainable operation of the industry. As there is a cogent need for industries to reduce over-dependence on main supplies due to different water management policies regarding discharge of wastewater and the use of mains supply by industries, successful wastewater treatment systems are needed, with the standard of treatment required usually influenced by the location and use (UN Water 2014b).

The brick industry produces wastewater with high concentrations of suspended solids that are not readily removed by preliminary treatments, and this highly turbid wastewater represents a major environmental challenge. The effluent stream from clay brick production plants is coloured and highly turbid due to the different sand and clay particles present in the effluent. There has been limited study of the characterization of these effluents and optimization of a treatment scheme for effluents generated from a brick production plant. Industrial wastewater containing clay or kaolin particles are difficult to treat because the particles do not readily settle due to their colloidal size, anisotropic shape and permanent negatively-charged basal faces (Nasser & James 2007; Morad et al. 2011). The discharge of clay-laden wastewater and sludge also poses a serious problem due to the clay particles' high affinity for water, which makes dewatering difficult (McFarlene et al. 2005; Nasser & James 2006). In the brick manufacturing industry, a successful wastewater treatment system will then depend on the efficiency of separating small-sized solid particles (especially colloids) from the liquid, hence, coagulation-flocculation appears to be a suitable treatment method.

To date, coagulation-flocculation has been widely used in the treatment of a variety of turbid industrial wastewaters. Examples include: breweries (Simate 2015), paper and pulp (Dilek & Bese 2001), mining (Abdelaal 2004), printing ink (Gdoura et al. 2013), textile (Suresh et al. 2013), petrochemical (Zawawi et al. 2013), food (Khannous et al. 2011), metallurgy (Berradi et al. 2014; Lopez-Maldonado et al. 2014), and stone-cutting industries (Fahiminia et al. 2013). The process is also capable of removing metals that are of environmental and public health concern such as aluminium (Berube & Dorea 2008) and arsenic (Hering et al. 1997) from surface water for potable uses. Bacteria and other pathogens in water – attached to suspended solids or freely suspended – have also been removed using coagulation-flocculation process (Prakash et al. 2014). The wide application of this process in many industries is due to its simplicity and cost-effectiveness (Najafi & Movahed 2009). Regardless of the nature of the treated water sample and the overall applied treatment scheme, coagulation-flocculation is usually included, either as a pre- or post-treatment step in water and wastewater treatment. Effluents containing clay suspensions have also been extensively treated using several polymer-based coagulants and flocculants, with high efficiency (Petzold et al. 2003). Polyacrylamide (PAM) and its derivatives are among the most extensively used polymers, with wide applications in industries such as aggregate, mining, metallurgical and clay processing (Pillai 1990; Mortimer 1991; Besra et al. 2003). However, the selection of the proper coagulant type and dose at optimum pH and mixing speed for a specific wastewater are still a major problem associated with the method, since the characteristics of wastewater differ among industries.

However, the efficiency of a coagulation-flocculation process is highly dependent on many factors. These include coagulant or flocculant type and dosage or amount (Kim et al. 2001), pH of the water (Amirtharajah & O'Melia 1990; Kim et al. 2001; Najafi & Movahed 2009), mixing speed and time (Amirtharajah & O'Melia 1990; Galluzzo et al. 1999), temperature and retention time (Zhu et al. 2004), etc. To deal with the method effectively, optimisation techniques have already been proposed such as the one-factor-at-a-time (OFAT), the response surface method (RSM) and genetic algorithm (GA) techniques (Li & Dawood 2013). The OFAT is a simple jar test technique that involves changing each factor (flocculant type, dosage, pH, and mixing speed) one step at a time, while keeping the level of other factors constant. The level of each factor or combination of factors that gives the best response or desired result (for example, highest turbidity removal efficiency) is selected as the optimum treatment condition (Li & Dawood 2013).

This research will then aim to devise an optimal treatment technique for effluents generated from the brick industry. The main objective of this study is to optimize the coagulation-flocculation treatment process by employing OFAT in order to obtain water of high quality suitable for reutilization – to date, there is limited study on this area of research. Dosage, pH and mixing speed, together with flocculant type, were the variable factors to be optimized, while turbidity and suspended solids removal efficiencies were chosen as the main dependent output variables. This optimum condition can then be used on a larger scale in industries so that the water-intensive brick making industry becomes self-sufficient in terms of water supply, attains its sustainability targets, enhances cost saving and ensures strict environmental regulations for wastewater discharge.

The wastewater samples were collected directly from the effluent outlet of a clay brick production plant located in the South-East of England. The samples were transported to the laboratory, where they were stored in the refrigerator at 4 °C, and analysed within 48 hours of collection. Sampling was carried out twice every week for 16 weeks. The duration of the sampling provided a representative sample and data set for different weather conditions and production patterns on site. Special precautions were taken during sample collection, transport and preservation, as stated in Sections 1060B and 1060C of the Standard Methods for the Examination of Water and Wastewater (APHA et al. 2005). Total solids (TS) and total suspended solids (TSS) concentrations were determined according to the Standard Methods for the Examination of Water and Wastewater, 2540 (APHA et al. 2005). Total dissolved solids (TDS) was calculated as the difference between TS and TSS. The pH, temperature and electrical conductivity (EC) of the wastewater samples were measured using a HACH HQ40d multi-parameter meter (HACH Lange GmbH, Düsseldorf, Germany) while turbidity was measured using a HACH 2100Q portable microprocessor turbidity meter. The colour of the sample was noted by visual observation. Inorganic metals were analysed using a PerkinElmer Optima 5300 DV inductive coupling plasma-optical emission spectrophotometer (ICP-OES) (PerkinElmer, Waltham, MA, USA). Chemical oxygen demand (COD) of the samples was determined using a HACH COD reactor while sulphates and ammoniacal nitrogen (NH4-N) were determined using a HACH DR 2800 spectrophotometer. All parameter analyses were performed three times and the results were expressed as mean value.

The crude sample obtained directly from the exit pipe of the production plant contained a significant amount of clay particles in a slurry form. The samples were allowed to settle for 1 hr and the range and mean characteristics of the crude samples obtained after pre-sedimentation is summarized in Table 1 below.

All reagents used in this study were analytical grades. The flocculants used include cationic polyacrylamide (cPAM), slightly anionic or neutral polyacrylamide (nPAM), anionic polyacrylamide (aPAM), and poly diallyldimethylaluminiumchloride (PDADMAC), with characteristics shown in Table 2. The pH of the water samples was adjusted using 0.1 M hydrochloric acid (HCl) and 0.1 M sodium hydroxide (NaOH). Standard solutions of samples and the stock polymer solutions were prepared by dissolution in distilled water.

The coagulation-flocculation experiment was carried out using a flocculator with six paddles, at room temperature. The conventional multifactor experiment was employed whereby optimization was carried out by varying a single factor while keeping all other factors constant. This method is also called the OFAT (one-factor-at a-time) method. The Stuart illuminated-base flocculator SW1 (Cole-Parmer, Vernon Hills, IL, USA) was used for the jar test experiment. The illuminated base aids in flocs observation. The jar test experiment was divided into 4 phases:

• a.

  Experiment I: Screening and selection of optimum flocculant

• b.

  Experiment II: Determination of optimum flocculant dose

• c.

  Experiment III: Determination of optimum pH

• d.

  Experiment IV: Characterization of samples at optimal conditions using optimum flocculant type, dose and pH as determined in experiments I–III.

Four different flocculants, nPAM, aPAM, cPAM and PDADMAC, were screened for optimal removal of turbidity and TSS under the same experimental conditions, using the flocculator to simulate the coagulation-flocculation treatment process. 500 mL of the pre-treated samples, which had undergone sedimentation, were added to each of the six 1 L capacity beakers placed in a flocculator. The range of volumes of each flocculant used was from 0.5 to 3 mL and the corresponding concentrations (dosage) are shown in Table 3. The samples were fast-mixed at a speed of 200 rpm for 5 minutes, after which a slow mixing speed of 50 rpm was applied for 15 minutes at a constant pH of 8.16. The flocculated samples were allowed to settle for 30 minutes and characterized for turbidity and TSS. The flocculant with the optimum performance was selected for further optimization.

A range of flocculant dose ranging from 1–6 mg L⁻¹ of the optimal flocculant from experiment I, was initially applied, as shown in Table 4. The samples were treated under the same rapid mixing speed of 200 rpm for 5 minutes, slow mixing speed of 50 rpm for 15 minutes, with a post-flocculation settling time of 30 minutes and constant pH of 8.16. The samples were analyzed and the dose with the optimum removal efficiency for turbidity and TSS was selected for further optimization. This was classified as experiment IIa. A duplicate experiment was performed around the initial optimum dose range and the experiment classified as IIb as shown in Table 5.

The pH of the sample solution was varied from 6.5 to 9 using 0.1 M HCl and 0.1 M NaOH solutions. The optimum dose from experiment IIb was added to each of the six beakers and placed in the flocculator. The same mixing condition of 200 rpm for 5 minutes and 50 rpm for 15 minutes was applied. The samples were allowed to settle for 30 minutes and analysed for turbidity, EC, TSS, TS and TDS. The pH with the optimal turbidity and TSS removal efficiency was selected. A duplicate experiment was carried out to further optimize the pH within the range of initially selected value. The initial pH optimization was classified as experiment IIIa and the latter pH optimization as experiment IIIb. The values are summarized in Tables 6 and 7 below.

A fresh sample from the production plant was characterized and treated at optimal conditions, based on optimum flocculant type, dose and pH obtained from experiments I to III. The treated sample was characterized for turbidity, TSS, TS, TDS, COD, NH4-N and sulphates. Duplicate analysis was made under similar experimental conditions to confirm reproducibility of the experimental results.

The performance of the four polymer flocculants used in this research were compared based on their turbidity and solids removal efficiency. The results are summarized in Figures 1 and 2 below.

The four different flocculants tested showed relatively good performance and removal efficiency for selected parameters such as TSS and turbidity as shown in Figures 1 and 2 respectively. However, nPAM showed the highest efficiency at a minimum dose of 1 mg L⁻¹ for turbidity and TSS removal with approximately 98% removal efficiency as shown in Tables 8 and 9. PDADMAC gave the best performance, but at a higher dose of 6 mg L⁻¹. The samples were tested at different periods, hence, the variation in average raw sample turbidity and TSS.

It was expected that the negative charge of the clay suspension would be readily neutralized by the positive charge of the cationic polymers such as cPAM and PDADMAC; however, this result showed that the effect of charge neutralization is less predominant to bridging during the flocculation of clay suspensions.

This trend is buttressed by an earlier work of Palomino & Kim (2009), which stated that polymer bridging is a more effective bonding mechanism than ionic interaction (electrostatic attraction) for water samples with a very high solids content. Hence, the higher efficiency of nPAM over cPAM and PDADMAC at minimum dose of 1 mg L⁻¹, confirmed that molecular mass of polyacrylamide had a greater impact on clay suspensions than charge type. A higher molecular mass polymer and concentration implies more monomer units per single polymer chain, thus more polymer bridging, leading to better flocculation. This also supports earlier studies comparing charge type and molecular weights of flocculants (Clark et al. 1990; James & Nasser 2006). The observed lower efficiency of aPAM, which has similar molecular weight and ionicity as cPAM and PDADMAC, shows that the neutralization mechanism is more effective when molecular weights of flocculants are the same.

nPAM, which gave the best performance at minimum concentration of 1 mg L⁻¹, was further optimized for dose and the results are summarized in Figure 3 below. The observed trend means that at doses below 0.8 mg L⁻¹, flocculation is incomplete. Turbidity and TSS removal efficiency peaked at 0.8 mg L⁻¹ and at this point the surface of the particles are fully saturated, hence, any dose above 0.8 mg L⁻¹ (saturation point) indicates an excess, which would result in charge reversal and particles re-stabilization, thus the observed decline in efficiency after the optimum dose. The re-stabilization of the flocculated particles would lead to re-dispersion of the clay particles (Ebeling et al. 2005). Also, the decline in efficiency on addition of dose in excess of 0.8 mg L⁻¹ can be attributed to the fact that at 0.8 mg L⁻¹, the entire particle surface are saturated with polymer, thus no site is available for bridging with other particles leading to the ‘hair-ball effect’ (Wakeman & Tarleton 1999; Ebeling et al. 2005).

The pH was optimized within the range of 6.5–9 at constant dose of 0.8 mg L⁻¹ of nPAM and mixing speed of 200 rpm for 5 minutes and 50 rpm for 15 minutes. The optimum pH obtained was 7.47 with turbidity and TSS reduced to as low as 1.54 NTU and 1.00 mg L⁻¹ respectively as shown in Figure 4 below.

The pH was further optimized within the range of 7–8 at the same constant dose of 0.8 mg L⁻¹ of nPAM and constant mixing speed. The optimum pH range obtained was between 7.2 and 7.4, with turbidity and TSS reduced to 1.48 NTU and 1.00 mg L⁻¹ respectively.

The optimum pH obtained from this study conforms to the theory that the choice of flocculants changes from non-ionic to highly anionic as the pH of sample increases from 0 to 14 (Pillai 1990). The carboxylate groups (RCOOR’) present in flocculants primarily aid in extending the polymer chain to enhance bridging, while the primary function of the amide groups (RCONH2) is to enhance adsorption by hydrogen bonding (Pillai 1990). As pH increases, the carboxylate groups present in anionic flocculants are ionized (RCOO-), which increases its activity. Therefore, a slightly anionic flocculant would show its greatest activity at pH range of 6–8, whereas at a pH greater than 9.5, highly anionic flocculants are most active (Pillai 1990).

In general, the polymeric coagulants are not readily affected by pH change as shown in the close range of efficiency results for the different pHs in experiment III. In addition, the change in pH before and after flocculation was minimal, as shown in Tables 10 and 11, further laying credence to the negligible effect of pH on polymeric flocculants.

Slightly anionic polyacrylamide (nPAM) was found to be the most efficient flocculant for turbidity and suspended solids removal, with efficiencies of 99% and 98% respectively. The optimum pH range obtained based on turbidity and TSS was 7.2–7.4 for the wastewater sample and the optimum flocculant dosage based on turbidity and TSS removal was found to be 0.8 mg/L, as summarized in Tables 12 and 13 below:

The summary of the physico-chemical parameters measured in the treated water and the corresponding recommended WHO values (WHO 2011) for drinking water, shown in Table 14 below, indicates that the treated water met most drinking water guideline values. Hence, the treated water can be reliably substituted for the mains water in the production cycle based on these findings.

As seen in Table 14, further characterization of water samples treated at optimal conditions of pH, dose, mixing speed and time showed that the amount of inorganic metals present in the crude samples was reduced after flocculation. In addition, NH4-N in the crude sample after pre-sedimentation was reduced after flocculation. The sulphate and COD content was still within acceptable guideline values, though there was a slight increase during pH optimization. The increase can be attributed to the dissolution of more ions on acidification of the crude sample to attain the optimum pH range of 7.24 required for flocculation. In contrast, at a normal pH of 8.13, there was a slight reduction in COD and sulphates after flocculation, thus laying credence to the plausible effect of acidification on sulphate and COD content. The dissolution of ions on acidification also explains the slight increase observed in EC and TDS concentrations at pH of 7.24, as shown in Tables 15 and 16 below.

Despite the increase in EC at an optimum pH of 7.24, the mean value in the treated water (1.15 mS cm⁻¹) was still below the maximum limit of 2.5 mS cm⁻¹ (WHO 2011) for drinking water. In addition, the sulphate content in the treated water at both optimum and normal pH were below the drinking water standard of 250 mg L⁻¹ for sulphates (WHO 2011).

Slightly anionic polyacrylamide (nPAM) was found to be the most efficient flocculant for turbidity and TSS removal, with efficiencies of 99% and 98% respectively. Hence, the use of slightly anionic polyacrylamide could help the brick manufacturing industry to start reducing the amounts of freshwater used as it has proven to be effective at a low concentration of 0.8 mg L⁻¹ with mean turbidity and TSS concentration values as low as 1.53 NTU and 1.75 mg L⁻¹ respectively. At normal water sample pH, the resulting efficiency of the flocculant was also found to be very high for most of the parameters tested, with 97% TSS and turbidity removal efficiencies. This will therefore reduce the treatment time and cost of acidification for pH balance, if the flocculant is applied on an industrial scale. Treatment at normal pH can also be supported owing to the slight reduction in COD, sulphates, TDS and EC compared to results obtained at an optimum pH of 7.24. A simple treatment process flow comprising the conventional coagulation-flocculation process at the optimal conditions as well as filtration was proposed as a multiple-barrier approach to ensure resultant water of high quality that can be recycled to augment mains water in the brick production cycle.

The authors wish to thank Wienerberger UK Ltd and Knowledge Transfer Partnership (KTP) for funding this research project. The authors also wish to thank The Petroleum Technology Development Fund (PTDF), Nigeria, for their study sponsorship.