Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 Physicochemical characterization and evaluation of PGA bioflocculant in coagulation-flocculation and sedimentation processes Valquíria Campos*, Ana R.A.C. Fernandes, Thales. A.M. Medeiros, Erik L. Andrade Instituto de Ciência e Tecnologia de Sorocaba, Univ. Estadual Paulista, UNESP, Brazil A R T I C L E I N F O Article history: Received 11 May 2016 Received in revised form 12 July 2016 Accepted 11 August 2016 Available online 12 August 2016 Keywords: PGA bioflocculant Polyaluminum chloride Aluminum sulfate Salto de pirapora WTP São paulo A B S T R A C T This paper analyzesanalyses the effectiveness of three commercial products: aluminum sulfate, aluminum polychloride and polyglutamic acid-based bioflocculant, in water clarification and purification. Poly-g-glutamic acid (PGA), a naturally occurring anionic polymer, is already produced on an industrial scale. A conventional jar test apparatus was used for flocculation experiments. The tests were performed using raw water from the Salto de Pirapora Water Treatment Plant (WTP) in the state of São Paulo, Brazil. The structure and elemental composition of PGA bioflocculant, which was characterized by Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), X-ray diffraction (XRD), X-ray fluorescence spectrometry (XRF) and thermogravimetric analysis (TGA), suggests that it is basically inorganic in nature, with high contents of calcium sulfate (87%) and g-polyglutamic acid (5%), among other chemical compounds. The performance of PGA bioflocculant was superior to that of the other coagulants. In the dry season, 80 mg L�1 of PGa21Ca removed 89.74% of apparent color, and residual turbidity was 5.78 NTU. In the rainy season, 90 mg L�1 of PGa21Ca removed 96.89% of apparent color and residual turbidity was 1.93 NTU. ã 2016 Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect Journal of Environmental Chemical Engineering journal homepage: www.else vie r .com/ locat e/ jece 1. Introduction Conventional water treatment methods include coagulation, flocculation and sedimentation processes, usually followed by filtration and disinfection. Coagulation–flocculation is considered the most important process in surface water treatment, and it can be used as a pre-treatment, post-treatment, or even main treatment [1]. The main parameters used in the physical characterization of natural water are color, turbidity, the various fractions of solids levels, temperature, taste and odor. Although these are physical parameters, they provide valuable preliminary information to characterize the chemical quality of water, e.g., suspended solids levels (associated with turbidity) and concen- trations of dissolved solids (associated with color), organic solids (volatile) and mineral solids (fixed), which are the compounds that produce odor, among other things. The purpose of removing or reducing turbidity is to increase the efficiency of chlorination to disinfect water. In a water treatment plant (WTP) that uses the complete water cycle treatment, a decrease in process efficiency and in rapid-mix, flocculation and sedimentation operations causes clogging of the filters, which thus * Corresponding author. E-mail address: vcampos@sorocaba.unesp.br (V. Campos). http://dx.doi.org/10.1016/j.jece.2016.08.011 2213-3437/ã 2016 Elsevier Ltd. All rights reserved. require more frequent washing, and may lead to increased turbidity and less effective disinfection of the water supply. Hence, the parameters of rapid-mix, flocculation and sedimenta- tion markedly influence the overall removal efficiency of impuri- ties contained in raw water. Therefore, these parameters should not be simply adopted, but established based on laboratory studies or a pilot plant. The coagulation-flocculation process also enables the identification of potential products that can be used as coagulants for water treatment, some of which are readily available, such as aluminum and iron-based coagulants, while others are experimental coagulants, such as PGA bioflocculant. According to the literature, the performance of each coagulant may also vary, depending on the type of water to be treated. Coagulants are used extensively in the treatment of public water supply systems. The two most widely used coagulants are aluminum and iron salts [2]. Aluminum sulfate was commonly used in Brazil because of its low cost and high efficiency in the treatment of public water supply. However, aluminum is not biodegradable and high concentrations of this metal are harmful to health, possibly even aggravating disorders such as Alzheimer’s disease [3,4]. Currently, WTPs use prepolymerized aluminum salt, known as polyaluminum chloride (PAC). Due to its basicity, PAC generally releases less acid than other coagulants during hydroly- sis, causing little variation in the pH level of treated water. For this http://crossmark.crossref.org/dialog/?doi=10.1016/j.jece.2016.08.011&domain=pdf mailto:vcampos@sorocaba. http://dx.doi.org/10.1016/j.jece.2016.08.011 http://dx.doi.org/10.1016/j.jece.2016.08.011 http://www.sciencedirect.com/science/journal/22133437 www.elsevier.com/locate/jece Table 1 Chemical composition of PGa21Ca analyzed by X-ray fluorescence (in mass%). Sample SiO2 Al2O3 SO3 MgO CaO Na2O K2O Cl2O Fe2O3 TiO2 PGa21Ca 6.63 2.97 47.33 0.49 40.41 0.09 0.13 0.99 0.71 0.05 3754 V. Campos et al. / Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 reason, media coagulated by PAC require little or no use of pH adjustment products. In addition to requiring only very slight pH adjustment, PAC is efficient in destabilizing particles, thus favoring the rapid formation of large uniform flakes without the need for flocculating aids. Although aluminum sulfate and PAC present favorable characteristics as inorganic coagulants, chemical for- mulations containing aluminum reportedly have harmful effects [5] and commonly produce a large amount of hazardous sludge containing residual metals [6]. Thus, there is a great demand for the development of a coagulant that poses no risk to the environment or to human health. To this end, numerous polymer matrices have been proposed for the removal of contaminants from water and for the treatment of wastewater. Biodegradable flocculants could represent an alternative to the conventional flocculants currently in use. Reports are available on the application of bioflocculants such as chitosan, which may be used for both coagulation and flocculation [1,7]. Sellami et al. [8] evaluated cactus juice as a bioflocculant in the coagulation- flocculation process for industrial wastewater treatment. Campos et al. [9,10] polymerized bentonites to obtain ionene derivatives, combining polysoap properties with organophilic properties for possible application in the removal of organic pollutants. Poly-g-glutamic acid (PGA) is commercially available as PGa21Ca, a powder containing 14% of poly-g-glutamic acid, which is stabilized with minerals such as calcium sulfate [11]. The biopolymer g-polyglutamic acid (g-PGA) is anionic, water-soluble, biodegradable, biocompatible, edible and non-toxic to humans and the environment. Due to these special characteristics, it has been applied in the environmental sector as a coagulant agent. Several researchers have described the use of g-PGA in water treatment [12–15]. Shih and Van [16] stated that poly-g-glutamic acid is a polyamide flocculant and it is considered the best option because of its high yield, high flocculating activity and ability to flocculate a wide range of organic and inorganic compounds. The purpose of this study was to evaluate the efficiency of commercial coagulants based on coagulation, flocculation and sedimentation tests. Experimental series were performed using traditional products such as aluminum sulfate, polyaluminum chloride and polyglutamic acid-based PGA bioflocculant, a Japanese product described as effective in the decontamination of water for human consumption. Another aim of this study is to understand the behavior and physicochemical stability of PGa21Ca. 2. Material and methods The bioflocculant PGa21CaTM from Nippon Poly-Glu Co. Ltd. was supplied in powder form, with a recommendation for the use of 0.05–0.1 g L�1. The product’s typical flocculation time is 15 min at an initial turbidity of 100 NTU, resulting in a final turbidity of <1 NTU. The aluminum sulfate and aluminum chloride were supplied by the Salto de Pirapora WTP. Raw water was collected from the Salto de Pirapora WTP, because of its importance for the metropolitan area of Sorocaba. The sample was digested with concentrated nitric acid and heated to 100 �C in a block digester to determine the cations in the raw water, using inductively coupled plasma optical emission spec- troscopy (ICP-OES, Spectro Arcos) in radial-view mode. The chromatographic method (HPLC/IC model 430) was used for the simultaneous determination of anions. The PGa21Ca was characterized to determine its chemical composition, thermal stability, and morphology by means of X-ray fluorescence spectrometry (XRF), X-ray diffraction (XRD), elemen- tal analysis, infrared spectroscopy, thermal analysis and scanning electron microscopy. Before subjecting the PGa21Ca to physico- chemical characterization, it was sifted through a stainless steel sieve (Tyler, 200 mesh) with 0.075 mm sieve opening size, placed in stoppered flasks, and stored in a desiccator. The PGa21Ca was analyzed by energy dispersive X-ray fluorescence (EDXRF) using a PANalytical Epsilon 3 spectrometer. Basal spacings were determined by X-ray diffraction (XRD) in a Rigaku MiniFlex diffractometer, using CuKa radiation (l = 0.15416 nm) and a scintillation detector at 30 kV and 15 mA at a rotation of 5–80� at a scale of 2u, applying 0.02� steps at 0.6 s/step. The data were analyzed using Diffraction Plus software, integrated with JCPDS � International Center for Diffraction Data (ICDD) version 2001. The purpose of the elemental analysis was to identify elements commonly found in organic compounds, such as carbon, hydrogen and nitrogen, and was performed in a Perkin Elmer 2400 CHN elemental analyzer. The elemental analysis was performed in duplicate. The infrared spectrum of absorption was obtained by means of Fourier transform infrared spectroscopy (FTIR), using a Perkin Elmer Frontier FTIR spectrometer. The PGa21Ca was thermally analyzed in the temperature range of 10–1000 �C, applying a heating rate of 20 �C min�1. The samples were degraded in a TA Instruments 50 thermogravimetric analyzer (TGA 50) in an air oxidizing atmosphere and nitrogen atmosphere, using a nitrogen flow of 50 cm3min�1 and a synthetic air flow of 100 mL min�1. A sample of about 12 mg was used in all the experiments. Structural information about PGa21Ca was obtained using a conventional scanning electron microscope (SEM). The micrograph was magnified by up to 5000 X after coating the surface of the sample with Au. The analysis was performed using a scanning electron microscope (LEO 440i) coupled to an energy dispersive X-Ray spectrometer (EDS) (Oxford Si (Li) detector). 3. Results and discussion 3.1. Physicochemical characterization of PGa21Ca The specific surface area (SBET) of PGa21Ca was determined by the Brunauer- Emmet-Teller (BET) method through physical adsorption of N2 at low temperatures, using measurements in the relative pressure range (P/P0) of 0.05–0.1. The pore volume was calculated by the Barrett-Joyner-Halenda (BJH) method [17] and by means of the N2 adsorption isotherm. PGa21Ca was found to have a specific surface area of 1799 m2g�1 and a pore volume of 0.004 cm3 g�1. The physicochemical analysis of PGa21Ca was performed via EDXRF, which revealed the percentage of oxides in the commercial product (Table 1). The results were normalized and expressed as Na and U oxides, excluding light elements such as H, He, Li, Be, B, C, N, O, F, and Ne. The elemental analysis of PGa21Ca indicated CHN percentages in the composition of the commercial product, which are listed in Table 2. According to Ho et al. [18], an elemental analysis of g-polyglutamic acid typically shows 44.86% (C), 5.91% (H) and 10.49% (N), which is consistent with the composition calculated here as 46.51% (C), 5.43% (H) and 10.85% (N). The g-polyglutamic acid content in PGa21Ca was estimated considering that the formula of g-polyglutamic acid is (C5H7NO3)n, while that based only on the monomer is 46.51% (C), 5.46% (H),10.84% (N) and 37.17% (O). However, it is prudent to deduce the g-PGA content from the Table 2 CHN content of PGa21Ca. . Sample C/% H/% N/% CHN Total PGa21Ca 2.45 2.53 0.04 5.02 Table 3 Interlayer spacing dhkl on dried basis (nm). PGa21Ca d/I d/I d/I Gypsum 0.426/100 0.755/88.21 0.266/36.27 Anhydrite 0.349/100 0.285/60.16 0.220/38.88 Note: d-spacing and I-intensity. V. Campos et al. / Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 3755 carbon analysis, based on the assumption that this element is present only in g-PGA, since hydrogen may be present in other salts or as hydration water and the nitrogen content was negligible. Owing to this low g-PGA content in the sample, the nitrogen results were not considered sufficiently reliable to deduce its content by the analytical combustion technique used here. In this case, considering the carbon content in g-PGA, the commercial product contained 5.26%. 3.2. X-ray diffraction The mineralogical composition of the sample was determined qualitatively based on an X-ray diffraction analysis. The XRD analysis showed peaks characteristic of gypsum CaSO4�2H2O and anhydrite Ca(SO4). The identified compounds are summarized in Fig. 1 and Table 3. No mineralogical species typical of silicates were found, but the chemical analysis revealed silicon, which may be present as amorphous silica. 3.3. PGa21Ca morphology The microstructure of the sample was examined by SEM. To this end, the powder sample was spread on an adhesive, double-stick carbon tape which was then pressed onto a 0.5 inch diameter stub. Various areas of the sample spread on the stub were examined, and after ensuring that its morphology and morphometry did not vary, images were recorded by a secondary electron detector. The SEM micrographs revealed morphologies typical of gypsite (Fig. 2). The mineral appeared in the form of thick platelets ranging in size from 10 to 30 mm, and at another moment, as aggregates of up to 30 mm in size, composed of particles of approximately one micron. The representativeness of the sample can be shown in other micrographs. Chemical mapping of its composition by EDS revealed the presence of Ca, S, Si and Al. Fig. 1. X-ray diffractog 3.4. Infrared spectroscopy The FTIR-ATR technique was used to identify the typical chemical groups of g-PGA in PGa21Ca, by means of the characteristic bands. The infrared spectra were obtained from FTIR grade potassium bromide pellets in the range of 4000– 400 cm�1 for observation of the functional groups (Fig. 3). The FTIR analysis of PGa21Ca in Fig. 3 shows vibrations characteristic of hydroxyl groups at a wavelength of 3400 cm�1, as well as vibrations in the range of 900 cm�1corresponding to the Al- OH2group. Standing out between the bands characteristic of aluminum sulfate are n1 at around 990 cm�1, n3 at around 1113– 1089 cm�1 and n4 at around 620–553 cm�1[19]. The position of the 3392 and 3495 cm�1 bands in some gypsum infrared spectra may occur close to 3400 and 3540 cm�1 [20]. One of the four vibrational modes of SO4 2� was observed in the spectrum of the PGa21Ca sample, in the n3 SO4 stretching mode, which occurs at around 1120 cm�1 for gypsum [21]. The gypsum showed two bands at 1685 and 1620 cm�1, which were attributed to the presence of two types of water molecules. The band at 1685 cm�1 is characteristic of weak molecular bonds, while the band at 1620 cm�1 is characteristic of strong molecular bonds [21]. The PGa21Ca spectrum showed bands characteristic of gypsum at 3400 cm�1. The gypsum also presented three absorption bands in the stretching region of the OH group (nOH) at 3245, 3392 and 3495 cm�1 [20]. The FTIR analysis also revealed vibration bands characteristic of g-PGA at wavelengths of 1620, 1113 and 675 cm�1. Ho et al. [18] described the FTIR absorption for calcium g-polyglutamate in KBr pellets as: Amide I, N��H bending band at 1622 cm�1; Amide II, stretching band not observed presumably due to strong ionic complexation; C¼O symmetric stretching band at 1412 cm�1; ram of PGa21Ca. Fig. 2. SEM images of PGa21Ca (500, 1000, 2500 and 5000X magnification). Fig. 3. Infrared spectra of PGa21Ca absorption bands in the range of 400–4000 cm�1. 3756 V. Campos et al. / Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 V. Campos et al. / Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 3757 C��N stretching band at 1116 cm�1; N��H bending band at 669 cm�1; and O��H stretching band at 3415 cm�1. 3.5. Thermal degradation of PGa21Ca PGA was analyzed by TGA/DTG and DSC thermoanalytical techniques in synthetic air and N2 atmosphere at a heating rate of 20 �C min�1 (Fig. 4). The TGA/DTG curves indicated that the sample was thermally stable up to 97.93 �C and decomposed in three main stages, between 152.4, 334.2 and 801.3 �C. The three well-defined thermal events in both synthetic air and nitrogen atmospheres were clearly represented by successive reactions involving thermal characteristics of calcium sulfate and aluminum sulfate. In the case of PGa21Ca, the thermogravimetric analysis revealed a slight mass loss of 0.6192 mg, with a calculated residual mass of 89.10%, during the decomposition process. Çilgi and Cetişli [22] stated that the thermal decomposition of aluminum sulfate hydrate occurs in four major stages, all endothermic. The peak temperatures for the dehydration reaction occur in Stage I (51.85-80.85 �C) and II (80.85- 184.85 �C). The aluminum sulfate decomposition step begins at 691 �C. Gypsum showed two peaks in the range of 150–200 �C, representing the stepwise removal of water. At low concentrations, gypsum yielded a single peak at about 150–160 �C. Free water had an endothermic effect at 100 �C. The hexagonal phase exhibited an endothermic peak at about 160–175 �C, the cubic form of gypsum at 300–325 �C [23]. The DTG curve of PGa21Ca showed well- defined thermal events characteristic of gypsum decomposition at 152.7 and 178.3 �C. The decomposition of gypsum occurs in two stages, with the formation of hemihydrate at around 130 �C and of anhydrite at around 170 �C. CaSO4:2H2O ! CaSO4:1=2H2O þ 1:5H2O ! 2CaSO4 þ H2O Ho et al. [18] made a thermal analysis of calcium g-poly- glutamate and found 20% hydrated water, a dehydration tempera- ture of 110 �C, and a decomposition temperature of 335.7 �C, but did not observe the melting point. Fig. 4. TG/DTG and DSC curves for PGa21Ca measured at a hea 3.6. Coagulation-flocculation-sedimentation tests One of the approaches of this study focused on coagulation- flocculation-sedimentation experiments of raw water samples from the Salto de Pirapora WTP (Latitude 23�39022,55” S and Longitude 47�34027,66” W), in the state of São Paulo, Brazil. The raw water analyses were performed no more than 24 h after collecting the samples. Sampling was carried out in 2015 during rainy and dry periods. A sample was frozen immediately after filtering to determine nitrogen and phosphorus compounds. Filtration was performed using a vacuum pump and hydrophilic membrane filters (Millipore) with 0.45 mm pores. Parameters such as alkalinity, anions, cations and trace metals were also analyzed, and are described in Tables 4 and 5. The geology of the municipality of Salto de Pirapora favors mining of limestone rock belonging to the São Roque Group, in the form of limestone lenses, and dolomitic phyllite and calc schist. This lithological substrate, allied to the region’s mining operations, leads to substantial calcium carbonate content in the raw water. The commercial coagulants used in the coagulation-floccula- tion-sedimentation tests were aluminum sulfate (16% Al2O3) and PAC (9.59% Al2O3), as well as PGa21Ca imported from Japan. The levels of coagulants were established based on preliminary tests and recommended by the WTP, but the parameters of pH and temperature of the raw water were not changed. The optimum operating conditions for bioflocculant treatment were determined by the jar test procedure and, in this case, the PGa21Ca was used directly in its powder form for the jar test. The experiments were performed on a laboratory scale (jar testing), whose main purpose was to vary the velocity gradient during the simultaneous stirring of six square 2-l acrylic test jars. In these jar test experiments, we applied rapid mixing times of 10 s, an average rapid mixing velocity gradient of 1000 s�1, flocculation time of 30 min, and an average flocculation velocity gradient of 15 s�1. Lastly, we studied the influence of the settling velocity of 1 cm min�1 with simultaneous collection of samples from the sampling port located 7 cm below the 2 L marking on the six jars. ting rate of 20 �C min�1 (nitrogen and air atmospheres). Table 4 Analytical results of raw water from the Salto de Pirapora WTP pertaining to the main dissolved chemical species (cations and anions), in mg L�1. Sample Na+ Ca2+ K+ Mg2+ Al Fe Mn Cl� HCO3 � NO3 � SO4 2� PO4 3� Dry season 8.56 20.11 4.62 2.43 0.19 0.23 LQ 14.77 54.68 3.72 7.49 0.13 Rainy season 5.63 12.95 4.54 2.70 16.29 5.77 LQ 9.06 45.32 3.26 7.12 LOQ Note: results below the limit of quantification of the method (LOQ = 0.01 mg L�1). Table 5 Physicochemical parameters of raw water from the Salto de Pirapora WTP in the dry and rainy seasons. Parameters Unit Dry Rainy pH – 7.50 6.92 color HU 720 750 turbidity NTU 458 633 alkalinity mg L�1 42 43 EC mS cm�1 157.9 260.17 T �C 25 24.8 Note: Hu, Hazen units; EC, electrical conductivity; NTU, nephelometric turbidity unit; T, temperature. 3758 V. Campos et al. / Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 After the tests, the residual turbidity and percent removal of apparent color of the best settling results (1 cm min�1) were calculated, as illustrated in Figs. 5–7 . Fig. 8 shows the results of the tests performed on water collected during the rainy season, and in this case, using only the commercial product with the highest efficiency in the previous phase, i.e., PGa21Ca. Fig. 5 shows the residual turbidity of 90.55% and the percent removal of apparent color of 9.67 NTU obtained with PAC, while Fig. 6 shows the residual turbidity of 85.49% and percent removal of apparent color of 12.61 NTU obtained with aluminum sulfate. The best values obtained with PAC were at the level of 90 mg L�1. The use of PAC in place of aluminum sulfate reportedly offers economic and environmental advantages because it significantly reduces neutralizing agents such as hydrated lime and sodium carbonate. PAC also promotes coagulation in any pH range and can be applied at two to eight times lower levels than aluminum sulfate, generating less sludge, providing better water clarification, and facilitating filtration by reducing the filter cleaning frequency [24,25]. Fig. 7 illustrates the use of PGa21Ca at a level of 80 mg L�1, with removal of apparent color and residual turbidity of 89.74% and 5.78 NTU, respectively. Note that PGa21Ca was effective even at a low level of 55 mg L�1 (81.24% and 11.81 NTU), approaching or even surpassing the results of residual turbidity of the other coagulants applied at higher levels in this study. 0 5 10 15 20 25 30 0 10 20 30 40 50 60 70 80 90 100 50 60 70 80 90 100 R es id ua l t ur bi di ty ( N T U ) % A pp ar en t co lo r re m ov al ( H U ) Coagulant dosage (mg L-¹) Color Turb idit y Fig. 5. Residual turbidity and apparent color removal rate of water treated with PAC (Vs 1 cm min�1) in the dry season. According to Pan et al. [26], the flocculation efficiency of glutamic acid is dosage-dependent. Research has shown that, in kaolin suspensions, g-PGA shows a flocculation rate of 95.8%, in the conditions of 300 mg L�1 of PGA, pH 7, a temperature of 25 �C and a 350 rpm agitation rate. Carvajal-Zarrabal et al. [11], who studied the use of PGA in the treatment of tequila vinasse, a waste generated in ethanol distilleries, reported that the use of g-PGA (250–300 mg L�1) combined with sodium hypochlorite and sand filtration successfully removed about 70% of turbidity and reduced the chemical oxygen demand (COD) by 79.5%, with the additional benefit of color removal. The authors, who called PGa21Ca a harmless biodegradable biopolymer flocculant, report the effec- tiveness of the product, due to the high turbidity removal rate even at the natural pH of the vinasse (3.5), highlighting its remarkable effect in removing turbidity. Fig. 8 shows variations in residual turbidity of 14.67–1.98 NTU and apparent color removal rates of 86.65–96.95% at levels of 60– 100 mg L�1. As can be seen, at levels starting from 70 mgL�1, turbidity was reduced to below 5 NTU, the maximum value permitted for drinking water in Brazil [27], while a level of 90 mgL�1 resulted in residual turbidity of 1.93 NTU (removal of 99.74%) and apparent color of 96.89%. The results indicate the satisfactory performance of PGa21Ca in the removal of apparent color and residual turbidity in the treatment of raw water in the rainy season, demonstrating flexibility when varying the water quality parameters. The chemical analysis of the composition of PGa21Ca indicated the presence of aluminum sulfate and g-PGA. The former is responsible for agglomerating suspended or colloidal impurities and the latter for forming bonds between the colloids, leading to the formation of denser particles and heavier flakes. According to Huang [28], chemical coagulants such as aluminum salts, iron salts, polyaluminum chloride, polyferric chloride and polyacrylamide are still widely applied in the treatment of drinking water due to their outstanding coagulation performance and low cost [29]. However, these common coagulants still induce several health- related and environmental problems. The potential neurotoxic hazard posed by aluminum is generally accepted [30], as is the genetic toxicity of acrylamide monomer [31]. 0 5 10 15 20 25 30 0 10 20 30 40 50 60 70 80 90 100 50 60 70 80 90 100 R es id ua l t ur bi di ty ( N T U ) % A pp ar en t co lo r re m ov al ( H U ) Coagulant dosage (mg L-¹) Color Turb idit y Fig. 6. Residual turbidity and apparent color removal rate of water treated with aluminum sulfate (Vs 1 cm min�1) in the dry season. 0 5 10 15 20 25 30 0 10 20 30 40 50 60 70 80 90 100 55 60 65 70 75 80 R es id ua l t ur bi di ty ( N T U ) % A pp ar en t co lo r re m ov al ( H U ) Coagulant dosage (mg L-1) Color Turb idit y Fig. 7. Residual turbidity and apparent color removal rate of water treated with PGa21Ca (Vs 1 cm min�1) in the dry season. 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 80 90 100 50 60 70 80 90 100 R es id ua l t ur bi di ty ( N T U ) % A pp ar en t co lo r re m ov al ( H U ) Coagulant dosage (mg L-¹) Color Turbidit y Fig. 8. Residual turbidity and apparent color removal rate of water treated with PGa21Ca (Vs 1 cm min�1) in the rainy season. V. Campos et al. / Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 3759 The excellent performance PGa21Ca in the removal of turbidity and apparent color is associated with the functions of its components, i.e., the inorganic fraction neutralizes colloidal charges owing to the presence of Ca2+ and Al3+ cations [32], and g-PGA acts as a chelator, forming bridges between the small particles. Table 1 shows the presence of aluminum in the formulation of PGa21Ca. However, the ability to form residual aluminum in water treated with PGa21Ca can be considered low, since g-PGA can bind to multiple metal ions, such as Al3+ [32]. 4. Conclusion Conventional water treatment tests were carried out using the jar test with six square beakers (2 L), which provided velocity gradients in the range of 10–1200 s�1. PAC showed better performance than aluminum sulfate in the removal of apparent color and turbidity from raw water from Salto de Pirapora. However, PGa21Ca, whose formulation contains an inorganic portion and another of y-PGA, showed better results in the removal of apparent color and turbidity from raw water than the other coagulants, and also caused little change in the water pH level. At 90 mg L�1, PAC presented an apparent color removal rate of 90.55% and residual turbidity of 9.67. In contrast, at an initial level of 90 mg L�1, aluminum sulfate removed 85.49% of apparent color while residual turbidity was 12.61 NTU. In the dry season, 80 mg L�1 of PGa21Ca removed 89.74% of apparent color and residual turbidity was 5.78 NTU. In the rainy season, at 90 mg L�1, PGa21Ca removed 96.89% of apparent color and residual turbidity was 1.93 NTU. PGa21Ca is basically composed of calcium sulfate (87%), g-polyglutamic acid (5.26%), and other chemical compounds. The chemical analysis of PGa21Ca indicated that the product is essentially inorganic. In fact, PGa21Ca can be considered a hybrid coagulant that combines aluminum sulfate and g-PGA. In this case, we understand that the presence of gypsum and anhydrite as excipients is justified since they facilitate dosing and also contribute to the formation of a water-insoluble compound, because g-polyglutamic acid can react with calcium, i.e., the solubility of CaSO4 in water is 2 g L�1, thus enabling it to react with g-PGA. In addition to containing calcium sulfate as excipient, the formulation of PGa21Ca includes a conventional coagulant, aluminum sulfate, calcium carbonate as the alkalizing agent, and g-polyglutamic acid as an adjuvant compound, i.e., an anionic polymer that is able to form bridges between coagulated particles and thus create larger flakes. In the process of coagulation-flocculation and separation of solids from the liquid phase by settling, PGa21Ca showed significant results in the removal of apparent color and in the residual turbidity of water samples collected in the dry and rainy seasons, especially with respect to the aluminum sulfate coagu- lants. However, it should be kept in mind that this product is inorganic and may produce large quantities of raw sludge, which would have to be managed. Acknowledgements This research was financed by the São Paulo Research Foundation � FAPESP (grant 2015/02650-8). The authors thank the Dean of Research of UNESP, as well as FAPESP, for granting science introduction scholarships, which are very helpful in our laboratory routine. We are also indebted to the Salto de Pirapora Water Treatment Plant of SABESP. References [1] L. Rizzo, G. Lofrano, M. Grassi, V. Belgiorno, Pre-treatment of olive mill wasterwater by chitosan coagulation and advanced oxidation processes, Sep. Purif. Technol. 63 (2008) 648–653. [2] T. Okuda, A.U. Baes, W. Nishijima, M. Okada, Improvement of extraction method of coagulation active components from Moringa oleífera seed, Water Res. 33 (15) (1999) 3373–3378. [3] W.J. Lukiw, Aluminum and gene transcription in the mammalian central nervous system – implications for Alzheimer's Disease, in: C. Exley (Ed.), Aluminum and Alzheimer’s Disease: The Science That Describes the Link, Elsevier Publishers, London, 2001, pp. 147–168. [4] P. Niquette, M. Frédéric, A. Azzouz, R. Hausler, Impacts of substituting aluminum-based coagulants in drinking water treatment, Water Qual. Res. J. 39 (3) (2004) 303–310. [5] M.R. Schetinger, C.D. Bonan, V.M. Morsch, D. Bohrer, L.M. Valentim, S.R. Rodrigues, Effects of aluminum sulfate on delta-aminolevulinate dehydratase from kidney, brain, and liver of adult mice, Braz. J. Med. Biol. Res. 32 (6) (1999) 761–766. [6] Y. Shi, M. Fan, R.C. Brown, S. Sung, J(Hans).V. Leeuwen, Comparison of corrosivity of polymeric sulfate ferric and ferric chloride as coagulants in water treatment, Chem. Eng. Process 43 (2004) 955–964. [7] D. Zeng, J. Wu, J.F. Kennedy, Application of a chitosan flocculant to water treatment, Carbohydr. Polym. 71 (2008) 135–139. [8] M. Sellami, Z. Zarai, M. Khadhraoui, Cactus juice as bioflocculant in the coagulation-flocculation process for industrial wastewater treatment: a comparative study with polyacrylamide, Water Sci. Technol. 70 (7) (2014) 1175–1181. [9] V. Campos, J. Dweck, C.A.O. Nascimento, C.M. Tcacenco, Thermal stability of ionene polymers, J. Therm. Anal. Calorim. 112 (2013) 1221–1229. [10] V. Campos, C.M. Tacenco, Synthesis of polycationic bentonite-ionene complexes and their benzene adsorption capacities, Polímeros 25 (2015) 146–153. [11] O. Carvajal-Zarrabal, C. Nolasco-Hipólito, D.M. Barradas-Dermitz, P.M. Hayward-Jones, M.G. Aguilar-Uscanga, K. Bujang, Treatment of vinasse from tequila production using polyglutamic acid, J. Environ. Manage. 95 (2012) 66–70. http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0005 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0005 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0005 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0010 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0010 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0010 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0015 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0015 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0015 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0015 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0020 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0020 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0020 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0025 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0025 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0025 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0025 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0030 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0030 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0030 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0035 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0035 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0040 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0040 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0040 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0040 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0045 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0045 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0050 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0050 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0050 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0055 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0055 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0055 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0055 3760 V. Campos et al. / Journal of Environmental Chemical Engineering 4 (2016) 3753–3760 [12] J.A.A. Arpal, Compuestos estequiométricos del ácido poli(y-glutâmico) com tensoactivos catiónicos, Dissertação, ETSEIB (2004). [13] M. Kunioka, Biodegradable water absorbent synthesized from bacterial poly (amino acid)s, Macromol. Biosci. 4 (2004) 324–329. [14] M. Taniguchi, K. Kato, A. Shimauchi, X. Ping, K.I. Fujita, T. Tanaka, Y. Tarui, E. Hirasawa, Physicochemical properties of cross-linked poly-c-glutamic acid and its flocculating activity against kaolin suspension, J. Biosci. Bioeng. 99 (2) (2005) 130–135. [15] B.C. Capelette, Emprego da quitosana como coagulante no tratamento de água contend Microcystis aeruginosa—Avaliação de eficiência e formação de trihalometanos, Dissertação, Universidade de Brasília, Brasília, 2011. [16] I.L. Shih, Y.T. Van, The production of poly-(y-glutamic acid) from microorganisms and its various applications, Bioresour. Technol. 79 (3) (2001) 207–225. [17] E.P. Barrett, L.G. Joyner, P.P. Halenda, The determination of pore volume and area distribuitions in porous substances. I. Computations from nitrogen isothermas, J. Am. Chem. Soc. 73 (1951) 373–380. [18] G.H. Ho, T.I. Ho, K.H. Hsieh, Y.C. Su, P.Y. Line, J. Yang, K.H. Yang, S.C. Yang, y- Polyglutamic acid produced by Bacillus subtilis (natto): structural characteristics, chemical properties and biological functionalities, J. Chin. Chem. Soc. Taip. 53 (2006) 1363–1384. [19] J.T. Kloprogge, R.L. Frost, Raman microscopy at 77 K of natural gypsum CaSO4.2H2O, J. Mater. Sci. Lett. 19 (2000) 229–231. [20] J.P. Soares, Estudo microestrutural do gesso-alfa produzido pelo processo hidrotérmico e calcinação a seco e sua influência nas propriedades mecânicas pós hidratação, Dissertação, Universidade Federal de Pernambuco, Recife, 2005. [21] K.P. Mandal, T.K. Mandal, Anion water in gypsum (CaSO4.2H2O) and hemihydrate (CaSO4.1/2H2O), Cem. Concr. Res. 32 (2) (2002) 313–316. [22] G.K. Çılgı, H. Cetişli, Thermal decomposition kinetics of aluminum sulfate hydrate, J. Therm. Anal. Calorim. 98 (2009) 855–861. [23] V.S. Ramachandran, M.R. Paroli, J.J. Beaudoin, A.H. Delgado, Gypsum and Gypsum Products in Handbook of Thermal Analysis of Construction Materials, Noyes Publications, Norwich, 2001. [24] M. Furtado, Coagulantes modificados e mais eficazes reforçam o poder da clarificação. http://www.quimica.com.br/agua-coagulantes-modificados-e- mais-eficazes-reforcam-o-poder-da-clarificacao/, 2009 (accessed 02.07.2016). [25] A. Matilainen, M. Vepsalainen, M. Sillanpaa, Natural organic matter removal by coagulation during drinking water treatment, Adv. Colloid Interface Sci. 159 (2010) 189–197. [26] Y. Pan, B. Shi, Y. Shang, Research on flocculation property of bioflocculant PG. a21 Ca, Modern Appl. Sci. 3 (6) (2009) 106–112. [27] BRASIL. Ministério da Saúde. Portaria 2.914, de 12 de dezembro de 2011. Dispõe sobre os procedimentos de controle e vigilância da qualidade da água para o consumo humano e seu padrão de potabilidade. Brasília: Ministério da Saúde, 2011 (accessed 02.07.16). [28] X. Huang, S. Sun, B. Gao, Q. Yue, Y. Wang, Q. Li, Coagulation behavior and floc properties of compound bioflocculant-polyaluminum chloride dual- coagulants and polymeric aluminum in low temperature surface water treatment, J. Environ. Sci. 30 (2015) 215–222. [29] Z. Li, S. Zhong, H.Y. Lei, R.W. Chen, Q. Yu, H.L. Li, Production of a novel bioflocculant by Bacillus licheniformis X14 and its application to low temperature drinking water treatment, Bioresour. Technol. 100 (14) (2009) 3650–3656. [30] S.C. Bondy, The neurotoxicity of environmental aluminum is still an issue, Neurotoxicology 31 (5) (2010) 575–581. [31] K.L. Dearfield, G.R. Douglas, U.H. Ehling, M.M. Moore, G.A. Sega, D.J. Brusick, Acrylamide: a review of its genotoxicity and an assessment of heritable genetic risk, Mutat. Res. Fund. Mol. Mech. Mutagen 330 (1995) 71–99. [32] I. Bajaj, R. Singhal, Poly (glutamic acid)—an emerging biopolymer of commercial interest, Bioresour. Technol. 102 (2011) 5551–5561. http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0060 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0060 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0065 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0065 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0070 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0070 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0070 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0070 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0075 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0075 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0075 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0080 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0080 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0080 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0085 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0085 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0085 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0090 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0090 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0090 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0090 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0095 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0095 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0100 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0100 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0100 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0100 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0105 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0105 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0110 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0110 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0115 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0115 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0115 http://www.quimica.com.br/agua-coagulantes-modificados-e-mais-eficazes-reforcam-o-poder-da-clarificacao/ http://www.quimica.com.br/agua-coagulantes-modificados-e-mais-eficazes-reforcam-o-poder-da-clarificacao/ http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0125 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0125 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0125 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0130 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0130 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0140 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0140 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0140 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0140 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0145 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0145 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0145 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0145 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0150 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0150 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0155 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0155 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0155 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0160 http://refhub.elsevier.com/S2213-3437(16)30297-4/sbref0160 Physicochemical characterization and evaluation of PGA bioflocculant in coagulation-flocculation and sedimentation processes 1 Introduction 2 Material and methods 3 Results and discussion 3.1 Physicochemical characterization of PGα21Ca 3.2 X-ray diffraction 3.3 PGα21Ca morphology 3.4 Infrared spectroscopy 3.5 Thermal degradation of PGα21Ca 3.6 Coagulation-flocculation-sedimentation tests 4 Conclusion Acknowledgements References