Purification of real car wash wastewater with complex coagulation/flocculation methods using polyaluminum chloride, polyelectrolyte, clay mineral and cationic surfactant

Finding solutions to the global ‘water problems’ is one of the biggest challenges of the 21st century (Smalley 2005), which requires actions such as promoting the more rigorous protection of water sources, the development of low-water technologies and novel water treatment methods and the maximization of used water reclamation. For the treatment of industrial wastewaters, special purification methods are required, since they usually contain different toxic and/or non-biodegradable contaminants like hydrocarbons, dyes, pesticides, etc. (Iqbal 2016; Abbas et al. 2018; Iqbal et al. 2019). Oily contaminants can be harmful to microorganisms, plants and animals; moreover, they can accumulate in the food chain and inflict various genotoxic, carcinogenic or mutagenic damage (Abdel-Shafy & Mansour 2016; Iqbal 2016; Ukpaka & Wami 2017; Abbas et al. 2018; Tasker et al. 2018). Oil contaminations can increase the number of degenerative diseases and decrease the life expectancy (Morounke et al. 2017; Johnston et al. 2019).

Car wash stations produce large and increasing volumes of wastewater, since the number of registered vehicles exceeded 1.2 billion in 2014 (worldwide), and it is estimated to be 2.0 billion until 2035 (Zhao et al. 2017; Currie 2018). Car wash wastewaters contain several pollutants such as hydrocarbons, oily pollutants, brake dust, detergents, surfactants, heavy metals, etc. (Jönsson & Jönsson 1995; Kiran et al. 2015). Efficient purification and reclamation of these waters are required both from environmental and economic reasons (Al-Odwani et al. 2007; Panizza & Cerisola 2010; Kiran et al. 2015). Moreover, the useable freshwater per vehicle is already limited in some countries, such as in the Netherlands, Scandinavian countries or in Australia (60–100 L/car), which makes the reclamation of used waters imperative (Kiran et al. 2015; Pinto et al. 2017).

There are several methods which can be used for the reclamation of car wash wastewaters, such as sand filtration (Al-Odwani et al. 2007; Zaneti et al. 2011), oil skimming (Al-Odwani et al. 2007), flotation (Zaneti et al. 2011), adsorption (Hamada & Miyazaki 2004), coagulation/flocculation (Zaneti et al. 2011; Etchepare et al. 2014; Mohamed et al. 2014; Rodriguez Boluarte et al. 2016), ozonation (Rodriguez Boluarte et al. 2016), electrochemical oxidation (Panizza & Cerisola 2010), biological treatments (Suzuki et al. 2000; Rodriguez Boluarte et al. 2016), electrocoagulation (Mohammadi et al. 2017) and membrane separation (Hamada & Miyazaki 2004; Li et al. 2007; Boussu et al. 2008; Lau et al. 2013; Kiran et al. 2015; Pinto et al. 2017). Due to the complexity of car wash wastewaters, single methods usually are not efficient enough and/or the achievement of satisfactory purification efficiency often proves to be too expensive (Brown 2000; Rodriguez Boluarte et al. 2016). Membrane separation is one of the most promising techniques, which can result in excellent purification efficiency (Lau et al. 2013; Pinto et al. 2017). The main advantages of membrane filtration are its low cost, easy scalability and low energy consumption (Chelme-Ayala et al. 2009). However, a general drawback of membrane filtration is the inevitable fouling mechanism, which leads to flux reduction and membrane amortization, hence reduced lifetime and high replacement costs (Li et al. 2007; Boussu et al. 2008; Panizza & Cerisola 2010). Therefore, efficient pre-treatment of car wash wastewaters is necessary to protect the membrane and to slow down the fouling, thus achieving lower cost and higher flux. Among the contaminants of car wash wastewaters, finely dispersed floating materials and even more the emulsified oil droplets are responsible for the membrane fouling and flux reduction, since these droplets form blocks in the pores and hydrophobic cake layer on the surface (Veréb et al. 2017a). Conventional oil skimming and flotation can eliminate free and dispersed oil (d_{oil droplets} > 20 μm), but oil-in-water emulsions require more effective elimination methods (Gryta et al. 2001; Chakrabarty et al. 2008; Souza et al. 2016; Veréb et al. 2017a).

Although common coagulants such as iron and aluminum chlorides/sulfates are not efficient enough for the elimination of emulsified oil, but in our previous study (Veréb et al. 2017b) it was proved that the combination of specific coagulant and flocculant like polyaluminum chloride and anionic polyelectrolyte can be efficient for the destabilization of micro- and nanosized oil droplets. Therefore, the mentioned coagulant/flocculant combination might be useful for the purification of car wash wastewaters before membrane filtration, resulting in significant fouling reduction and longer lifetime of the membranes. In this study, purification of real car wash wastewater was investigated by the application of different coagulants/flocculants, including polyaluminum chloride, polyelectrolytes, and clay mineral. This latter material can intensify the sedimentation of the flocks, and its adsorption capacity also can be favorable to reach higher elimination efficiency of dissolved organic compounds (Zhu & Ma 2008; Szabo et al. 2011; Djehaf et al. 2017). Additionally, the effect of organophilic clay mineral – generated in situ by the addition of hexadecyltrimethyl ammonium bromide cationic surfactant – was also investigated. Since HTABr as an organic cation carrier can exchange the Na⁺ ions of clay minerals, thus it can make them suitable for the adsorption of different organic contaminants such as anionic (Shen et al. 2009) or cationic (Tonlé et al. 2008) dyes and different non-polar contaminants such as hydrocarbons (Wiles et al. 2005; Chikwe et al. 2018; Jennifer & Ifedi 2019), phenols (Senturk et al. 2009; Jennifer & Ifedi 2019) and naphthalene (Zhu et al. 2008). The main aim of the present study was to investigate the possibly beneficial combination of polyaluminum chloride, bentonite, a suitable polyelectrolyte and a cationic surfactant to achieve outstanding purification efficiency by a relatively simple coagulation/adsorption/sedimentation method in the case of real car wash wastewaters.

The purification of real car wash wastewater was investigated in a F4P Jar Test flocculator (VELP Scientifica). As coagulants polyaluminum chloride (‘BOPAC; Unichem Kft., technical grade), iron(III) chloride (‘UNIFLOC-C’, Unichem Kft., technical grade), iron(III) sulfate (‘UNIFLOC’, Unichem Kft., technical grade) or aluminum(III) chloride (‘UNIPAC’, Unichem Kft., technical grade) were used. As flocculants, non-ionic or anionic polyelectrolytes (‘UNIFLOC-M20’ and ‘UNIFLOC-LT27’, respectively; Unichem Kft., technical grades) were applied. Addition of clay mineral (Na-bentonite; Unikén Kft., technical grade) and a cationic surfactant – hexadecyltrimethyl ammonium bromide (‘HTABr’; Sigma-Aldrich, analytical grade) – was also investigated since organophilic clay minerals can adsorb different dyes and non-polar contaminants.

Coagulants/flocculants and other supporting materials were added in calculated amounts during the intense stirring (200 rpm) of 0.5 L car wash wastewater, and 30 seconds after the last material was added, 2 min slow stirring (20 rpm) was applied, then the formed flakes were allowed to sediment for 30 min. In the case of Na-bentonite addition, it was always added firstly in suspension form (before the experiments the bentonite was allowed to swell in stirred water for 48 h at minimum, c_suspension = 10 g/L). The HTABr – whose amounts were calculated according to the total cation exchange capacity – was added directly after the clay mineral and the suspension was left to be stirred vigorously for 20 min (200 rpm) to enable the complete exchange of Na⁺ ions of the clay mineral with the cationic surfactant. After 20 min of stirring, the coagulant and flocculant were also added to the beakers, then 2 min of gentle stirring (20 rpm) was applied. The volumes of the produced sediments were determined by Imhoff cone sedimentation experiments.

Purification efficiencies were determined by measuring the turbidity of the supernatants after 30 min of sedimentation with a Hach 2100N nephelometric turbidity meter. Additionally, chemical oxygen demand (COD) and extractable oil content were also measured in some cases. COD was measured by the standard potassium dichromate oxidation method using standard test vials (Hanna Instruments) applying 120-min-long digestions at 150 °C in a Lovibond ET 108 digester, then the measurements were carried out via a Lovibond COD Vario photometer. Extractable oil content was measured by a Wilks InfraCal TOG/TPH analyzer, using hexane (VWR International Ltd, analytical grade) as extracting solvent.

The car wash wastewater was collected from a South Hungarian car wash station and the results of its characterization were the following: COD was 357 ± 8 mg L⁻¹, conductivity was 1.43 ± 0.1 mS cm⁻¹, turbidity was 333 ± 10 NTU, extractable oil content was 26 ± 2 mg L⁻¹, and the pH was 7.30 ± 0.2.

In the first series of experiments, different amounts of BOPAC coagulant were added to the wastewater and the achievable purification efficiencies were determined by turbidity measurements. The sediment volumes were also measured (Figure 1).

The turbidity values of the supernatants decreased intensely in accordance with the increasing concentration of Al³⁺ (from 2 to 20 mg L⁻¹), resulting in 35 ± 4 NTU and 89 ± 1% purification efficiency, but higher concentrations resulted in only slightly lower turbidity values. At the same time, the volume of the produced sediment showed a nearly linear increase with the increasing Al³⁺ content. Considering the purification efficiencies, the quantity of the used chemicals and the produced sediment volumes, the 20 mg L⁻¹ Al³⁺ concentration could be recommended. 50 mg L⁻¹ aluminum content resulted in just a slightly higher purification efficiency (93 ± 1%), but twice higher sediment volume compared to the observed values in the case of 20 mg L⁻¹ added aluminum content.

In the next experimental series, different amounts of Na-bentonite (0, 100, 250, 500 mg L⁻¹) were added before the BOPAC addition (20 mg L⁻¹ Al³⁺ content was applied). The purification efficiencies (calculated from turbidity) and the sediment volumes are summarized in Figure 2(a).

100 mg L⁻¹ Na-bentonite addition significantly increased the purification efficiency (98 ± 1%) and the sediment volume decreased slightly. Following the application of higher bentonite amounts, almost the same purification efficiencies were measured and the sediment volume decreased only slightly. Therefore, higher than 100 mg L⁻¹ Na-bentonite concentration cannot be recommended, even though the sedimentation was slightly faster at higher bentonite concentrations as it can be seen in Figure 2(b). The effects of the sole addition of Na-bentonite was also investigated as reference experiments, and the measured turbidity values (333, 210, 285, 355 and 580 NTU in the cases of 0, 100, 250, 500 and 1,000 mg L⁻¹ bentonite additions, respectively) proved the necessity of the coagulant addition.

The extra addition of two different types of polyelectrolytes – after the Na-bentonite (100 mg L⁻¹) and BOPAC (20 mg L⁻¹ Al³⁺) additions – was investigated in two beakers, while in another two beakers reference experiments were carried out by using the polyelectrolytes alone. The non-ionic (UNIFLOC-M20) and anionic (UNIFLOC-LT27) polyelectrolytes were applied in a low concentration (0.5 mg L⁻¹). The results (Table 1) confirmed that the extra addition of both polyelectrolytes increased the purification efficiency, but concerning the sediment volume the anionic UNIFLOC-LT27 was more beneficial, as the measured sediment volume (V = 15 mL) was only half the value compared to the one measured after the addition of non-ionic polyelectrolyte (V = 30 mL).

Polyaluminum chloride (BOPAC) was compared with different conventional coagulants (iron(III) chloride, iron(III) sulfate and aluminum(III) chloride) to determine the achievable turbidities. Coagulants alone (used in 20 mg L⁻¹ metal ion concentrations), and together with the anionic polyelectrolyte (0.5 mg L⁻¹) and/or Na-bentonite (100 mg L⁻¹) were also investigated (Table 2).

The results confirmed the outstanding efficiency of polyaluminum chloride in the case of the real car wash wastewater, as the measured turbidity values were 4–35 NTU, while in the case of conventional iron or aluminum coagulants 88–190 NTU and 22–82 NTU values were measured, respectively. The outstanding purification efficiency of the BOPAC can be explained by its pre-hydrolyzed form and Keggin structure, which enable its advanced adsorption ability. In the series where BOPAC was added to the system, the COD values and extractable oil contents of the supernatants were also measured, and 57 ± 1% and 83 ± 3% purification efficiencies were determined, respectively.

The effect of in situ addition of HTABr was also investigated, since it can exchange the Na⁺ ions of bentonites to organic cations, hence it can modify the surface to be organophilic, making them suitable for the adsorption of different types of organics such as dyes and different non-polar contaminants. Different amounts of Na-bentonite (0, 100, 500 and 2,000 mg L⁻¹) and HTABr (in calculated concentrations, according to the bentonite's cationic exchange capacity) were added before the BOPAC (20 mg L⁻¹ Al³⁺) and UNIFLOC-LT27 (0.5 mg L⁻¹) addition. Turbidity, COD and extractable oil content of the supernatants and the sediment volumes were measured (Table 3).

The reduction of turbidity, COD and extractable oil content values were very similar in all cases, but enhanced color removal of the supernatants were observed visually when the organophilic bentonite was used. Absorbance measurements confirmed the increasing color removals from sample 1 to sample 4, as the absorbance values significantly decreased at λ = 400 nm (Figure 3(a)).

Considering the turbidity values, the color removal, and the reduced sediment volumes (Table 3), the usage of 500 mg L⁻¹ organophilic bentonite can be considered advantageous. The beneficial effect of HTABr addition via the formation of more compact flocks (which resulted in reduced sediment volumes) can be seen in Figure 3(b), where the treated waters were stirred slowly (20 rpm).

Similarly to the research of Rodriguez Boluarte et al. (Rodriguez Boluarte et al. 2016) the present study also proved that the usage of polyaluminum chloride is more effective than conventional coagulants like aluminum chloride, or iron-salts in the case of the chemical destabilization of real car wash wastewaters. The mentioned authors did reduce the turbidity to a very similar value (∼4 NTU), but polyaluminum chloride was used in a significantly higher concentration (500 mg/L) without any polyelectrolyte, bentonite, or cationic surfactant addition. Therefore, the treated wastewater was still colorful which was then treated by an additional ozonation step, but it was not able to further reduce the turbidity or COD of the wastewater. In the present study, turbidity, COD and color were also significantly reduced by a simple and cost-effective method. The achieved turbidity (4 NTU) was significantly lower compared to other studies: 12.5 NTU (Zaneti et al. 2011), 13–28 NTU (Etchepare et al. 2014) and 22 NTU (Mohamed et al. 2014). Simultaneously, the total cost of the presented purification method was calculated to be only 0.092 $ m⁻³ in the case of BOPAC (20 mg L⁻¹ Al³⁺), anionic polyelectrolyte (0.5 mg L⁻¹) and bentonite (100 mg L⁻¹) addition, and it was 0.372 $ m⁻³ when HTABr addition was also applied with 500 mg L⁻¹ bentonite in the case of the same BOPAC and polyelectrolyte addition. These costs are significantly lower compared to the calculated costs of other publications, which were e.g. 0.85–1.12 $ m⁻³ (Etchepare et al. 2014) and 0.812 $ m⁻³ (Mohammadi et al. 2017).

In the present study, it was demonstrated that the usage of BOPAC by itself (without Na-bentonite or polyelectrolyte addition) was not effective enough for the purification of real car wash wastewaters. The extra addition of anionic polyelectrolyte was crucial to significantly increase the size of the produced clusters, intensify the sedimentation and decrease the sediment volume. BOPAC proved to be significantly more effective than conventional coagulants due to its pre-hydrolyzed form and Keggin structure. Taking the efficacy and economical aspects into account the usage of 100 mg L⁻¹ Na-bentonite with 20 mg L⁻¹ Al³⁺ and 0.5 mg L⁻¹ anionic polyelectrolyte were the most effective to decrease the turbidity, COD and the extractable oil content of real car wash wastewaters with efficiencies of 98%, 59%, and 85%, respectively. These results reinforced that, by the combination of these materials, high purification efficiency can be achieved via a very simple method, without high investment cost. Simultaneous addition of the cationic surfactant (HTABr) resulted in the successful in-situ generation of organophilic clay minerals which resulted in significant color removal, slightly lower turbidity value (99% efficiency) and significantly lower sediment volume. The application of both these last two combinations can be recommended as easily feasible, but efficient pretreatments of car wash wastewaters before the final membrane filtration step for the efficient removal of floating particles, emulsified oil, and hydrocarbons.

The authors are grateful for the financial support of the Hungarian Science and Research Foundation (2017-2.3.7-TÉT-IN-2017-00016), the Hungarian State and the European Union (EFOP-3.6.2-16-2017-00010). G. V. thanks the support of the János Bolyai Research Scholarship of the Hungarian Academy of Sciences and the New National Excellence Program of the Ministry of Human Capacities (UNKP-18-4-SZTE-78). E.N.S. was supported by the Stipendium Hungaricum Scholarship Program. The authors are grateful to T. Gyulavári for his valuable contribution in proofreading the manuscript.