Membrane bio-reactor for advanced textile wastewater treatment and reuse

Transcription

1 Membrane bio-reactor for advanced textile wastewater treatment and reuse Dipartimento di Ingegneria Civile, Università di Firenze, Via S. Marta 3, Florence, Italy ( Abstract Textile wastewater contains slowly- or non-biodegradable organic substances whose removal or transformation calls for advanced tertiary treatments downstream Activated Sludge Treatment Plants (ASTP). This work is focused on the treatment of textile industry wastewater using Membrane Bio-reactor (MBR) technology. An experimental activity was carried out at the Baciacavallo Wastewater Treatment Plant (WWTP) (Prato, Italy) to verify the efficiency of a pilot-scale MBR for the treatment of municipal wastewater, in which textile industry wastewater predominates. In the Baciacavallo WWTP the biological section is followed by a coagulation flocculation treatment and ozonation. During the 5 months experimental period, the pilot-scale MBR proved to be very effective for wastewater reclamation. On average, removal efficiency of the pilot plant (93% for COD, 96% for ammonium and 99% for total suspended solids) was higher than the WWTP ones. Color was removed as in the WWTP. Anionic surfactants removal of pilot plant and WWTP were very similar (92.5 and 93.3% respectively), while the non-ionic surfactants removal was higher in the pilot plant (99.2 vs. 97.1). In conclusion the MBR technology demonstrated to be effective for textile wastewater reclamation, leading both to an improvement of pollutants removal and to a draw-plate simplification. Keywords Color removal; industrial wastewaters; membrane bio-reactors; membrane treatments; surfactants removal; textile wastewater Introduction Textile wastewater overview Textile industries carry out several fiber treatments using variable quantities of water, from five to forty times the fiber weight, and consequently generate large volumes of wastewater to be disposed of. such treatments include dyeing preliminary treatments (bleaching, desizing, mercerization), textile ennobling treatments (from dyeing to post-dyeing treatments, such as those required to increase colorant fastness, in wet and dry conditions) and then finishing, including operations such as fulling or impregnation with products giving special characteristics to fibers. Upon treatment completion, the bath has to be discharged. In this way, the effluent is enriched with compounds having high environmental impact and difficult to treat directly, through conventional biological processes. Table 1 shows some of the qualitative characteristics typical of textile wastewater. This article depicts the outcomes of a research aiming at verifying MBR technology applicability for the treatment of wastewater coming from Prato textile industrial districts (Tuscany). This study is part of a larger research framework whose target is to identify the most appropriate technologies to boost reuse of purified waters for industrial purposes, in the area of Prato, and in the well-watered Pistoia nursery district. Materials and methods The Baciacavallo WWTP The Baciacavallo plant is the main WWTP in the area of Prato, its capacity is of about 750,000 p.e. and has a maximum flow capacity of 6,000 m 3 /h. Most of waste waters flowing into the plant have textile industry origin (about 80%), whilst the remaining 20% comes Water Science and Technology Vol 50 No 2 pp IWA Publishing

2 Table 1 Effluent characteristics from the textile industries (Water Research Commission of South Africa, 2000) Process Composition Nature Sizing Starch, waxes, carboxymethylcellulose, High in BOD and COD polyvinyl alcohol Desizing Starch, glucose, carboxymethylcellulose, High in BOD and COD, suspended and polyvinyl alcohol, fats and waxes dissolved solids Scouring Caustic soda, waxes, grease, soda ash, Dark coloured, high ph, high COD, dissolved sodium silicate, fibres, surfactants, sodium solids phosphate Bleaching Hypochlorite, chlorine, caustic soda, Alkaline, suspended solids hydrogen peroxide, acids, surfactants, sodium phosphate Mercerising Caustic soda High ph, low COD, high dissolved solids Dyeing Various dyes, mordants, reducing agents, Strongly coloured, high COD, dissolved solids, acetic acid, soap low suspended, solids, heavy metals Printing Pastes, stanch, gums, oil, mordants, acids, Highly coloured, high COD, oily appearance, soaps suspended solids Finishing Inorganic salts, toxic compounds Slightly alkaline, low BOD from domestic buildings. As a consequence, inlet waters (about 120,000 m 3 /d on average) are characterized by the high concentrations of surfactants, textile oils and colorants. The treatment chain includes the following sections: Preliminary treatments: screening and degritting; Primary treatment: coagulation flocculation (with ferric chloride and organic polymers) and primary settling; Biological oxidation in carousel-type tanks; Secondary settling; Clariflocculation by adding aluminium chloride and anionic polyelctrolyte; Ozonation. The treated wastewater is partially reintroduced in the surface water system and partially (100 l/s) is further refined and reused to feed the industrial and fire-fighting waterworks of one of the main industrial areas in Prato. The refining treatment consists in sand filtration, biologically activated granular carbons and final chlorination. Main water chemical-physical values, at Baciacavallo WWTP inlet, are summarized in Table 2. The MBR pilot-scale plant The pilot-scale MBR (see Figure 1 for the flowchart), installed at the Baciacavallo WWTP, is made up of: a peristaltic pump for wastewater feeding; Table 2 Baciacavallo plant main inlet water chemical-physical parameters, measured on daily average samples Variable Media Max Min STD 114 Q (m 3 /d) 119, ,700 31, ,894 COD (mg/l) 686 1, TSS (mg/l) 228 1, N-total (mg/l) MBAS (mg/l) Non-ionic surfactants (mg/l) Absorbance at 420 nm

3 Figure 1 MBR pilot-scale plant flowchart a strainer for textile process residual fiber removal; a biological reactor with a total volumetric capacity of 3 m 3 ; a biomass ultrafiltration system equipped with 2 centrifugal pumps (Feed Pump, FP and Recirculation Pump, RP) and 1 filtering module of the plate and frame type (Pleiade, Rhodia-Orelis). The pilot-scale plant is part of the treatment chain, and operates in parallel with the oxidation-nitrification treatment of the Baciacavallo WWTP, therefore, downstream the coagulation flocculation primary settling phase. The biological reactor operates at constant level and the bio-mass is maintained in aerobic conditions via aeration, through 6 fine bubble diffusers. The ultra-filtration module (Rhodia, UFP10) is of the external type, with plate and frame membranes, where a cross-flow type filtration is performed. The module consists of 4 elements in series (each one made up of 7 membranes) for a total filtering surface of about 3 m 2. The membrane is of the organic type, made up of acrylonitrile copolymers, with a 3 µm thickness and a pore cut-off of 40 KD (approximately, this corresponds to a pore average size of µm). The average module inlet flow rate is of 30 m 3 /h, which turns into a cross-flow velocity of 2.1 m/s. Start-up and operation For the start-up, the bio-reactor was filled with 2,500 liters of activated sludge from the Baciacavallo WWTP. At the beginning of the filtering cycles, the inlet pressure was set to 1.8 bar; this made it possible to obtain the flow conditions specified by the manufacturer. The module was let free to self-regulate in terms of inlet pressure and permeate flow. When the permeate flow reached too low values (lower than 35 l/h/m 2 ) or the module inlet pressure exceeded 2.5 bar, a chemical cleaning was normally performed in order to restore the initial values of flux, transmembrane pressure and pressure drop. Membrane chemical cleaning, according to a procedure provided by the manufacturer, is performed according to the following steps: rinsing with permeate at ambient temperature for 15 min; washing at 40 C for 30 min with a sodium hypochlorite (1 g/l) and sodium hydroxide (5 g/l) water solution; rinsing through neutralization; washing at ambient temperature for 15 min with a nitric acid water solution (5 g/l); final rinsing through neutralization. As a consequence of permeate values obtained, the primary wastewater flow-rate at the 115

4 pilot-scale plant inlet, was comprised between 100 and 150 l/h, this resulted in an average COD volumetric load of 1.2 kgcod/m 3 /d. Monitoring and analytical methods The pilot-scale plant influent and permeate analysis were carried out with a minimum interval of three times per week; following elements were monitored: ph, total suspended solids, COD, nitrogen (as ammonium, nitrite and nitrate), MBAS surfactants, non-ionic surfactants, color (absorbance at 420 nm). Furthermore, a characterization of inlet wastewaters was conducted, making special reference to total COD fractionation in the soluble and particulate component, according to Mamais et al., (1993) methodology. As regards the bio-mass, the total and volatile suspended solid concentration was periodically determined, in addition to characterization through observations made with a microscope. Results and discussion COD Fractionation Tests conducted on COD fractionation, in its soluble and particulate components, provided a total COD average value of 869 mg/l, whose soluble component corresponded to 34.7%. Such value is rather moderate with respect to values found in other situations. As regards exclusively textile wastewater, Orhon (1998), reports a soluble COD fraction average value of 49% and Germirli Babuna et al. (1999) have found even higher values (about 70%). The soluble COD percentage lowering, can probably be ascribed to mixing of textile wastewater with domestic wastewater which may also determine a considerable contribution of particulate COD. Bio-mass development and characteristics The initial concentration of the biomass introduced in the bio-reactor was of 5 gtss/l. Fluctuation in solid particle concentrations, together with inlet wastewater characteristics variability, led the pilot plant bio-mass to operate with extremely variable organic loads. After an initial start-up period, the bio-mass grew with a linear trend until it reached about 16 gtss/l, in the space of 120 trial days. Permeate flow The trend of permeate flow extracted from the pilot plant was comprised between 35 and 65 l/h m 2, considerably lower than the 100 l/h m 2 specified by the manufacturer. During the experimental period, on the basis of the previously described criteria, four chemical cleanings of the module were required, actually with a monthly frequency. The cleaning system proved to be efficient in restoring the flow conditions. However, the permeate flow descending trends were not regular and this can be explained on the basis of following phenomena: substantial fluctuation, even unexpected, of solid particle concentrations in the aerated mixture, due to sludge escape; change in the sludge viscosity and filterability characteristics, qualitatively measured during observations made with a microscope, and in the analysis relevant to the determination of activated sludge solid particle concentration. 116 COD removal The COD value at the pilot plant inlet was comprised between 500 and 1,700 mg/l according to a substantially cyclic trend with a weekly step, clearly affected by changes associated with industrial activity. Figure 2 compares the COD trends of pilot plant permeate, Baciacavallo WWTP s secondary effluent and final effluent after ozonation treatment.

5 Figure 2 COD values at the pilot plant outlet and from the Baciacavallo plant (secondary and tertiary effluent) After about 2 weeks from start-up, a quick COD decrease at the outlet was observed; these values stabilized within 40 and 60 mg/l (mean value: 56.8 mg/l), in the third week, seldom exceeding the upper threshold. The removal efficiency, 93% on average, proved to be considerably higher with respect to that obtained only with biological treatment and secondary settling in the Baciacavallo WWTP. Moreover, also the pilot plant constant efficiency is noteworthy: outlet values were substantially independent from inlet loads. Nitrification Nitrification process results were extremely satisfactory. With respect to the full scale plant, the nitrification process efficiency appeared considerably higher. The nitrification process proved to be complete since no nitrite accumulations were found in the oxidation tank (all values were below the 0.05 mg/l threshold). Color removal The pilot plant operating efficiency in color removal was extremely interesting. Figure 3 shows weekly mean values (except Monday, Saturday and Sunday values) of the pilot plant influent and permeate absorbance; the same chart also shows the removal efficiency trend, always in terms of absorbance. Figure 3 Weekly absorbance mean values at 420 nm of Baciacavallo effluent (clariflocculation and ozone) and of the pilot plant permeate 117

6 With absorbance values comprised between 0.2 and 0.5, as far as permeate is concerned, values comprised between 0.04 and 0.1 were found, with percentage reductions ranging between 70 and 80%. Measured permeate absorbance values were significantly lower than the effluent values downstream the clariflocculation treatment, and comparable to those of the effluent from the ozone treatment. Such results can be explained with the increase in the absorption capacity by the bio-mass because of its destructuration, with respect to conventional sludge. This phenomenon is further stressed by the higher concentration of colorants in the oxidation tank, due to partial retention operated by the membranes. Finally, a partial biodegradation of colorants (difficult in an aerobic environment) possibly incomplete, and limited to the chromophore group simple breakage, cannot be excluded. Surfactants As a general rule, the pilot plant proved to be efficient in surfactant removal from textile wastewater; however, with respect to removal efficiency obtained in the full scale plant, the MBR technology effect appeared different between anionic and non-ionic compounds. Table 3 summarizes concentration mean values measured at the inlet, in the pilot-scale plant permeate and in the Baciacavallo plant effluent (downstream the secondary settling and the ozone treatment). With respect to the conventional biological treatment, therefore, an improvement in MBAS removal was noticed. In this case, the retaining action by membranes is deemed incidental, since in other studies (Marcucci et al., 2002) it was found that, by using UF membranes with 70 kda MWCO, therefore a situation quite similar to the one under examination, no significant reduction in the compounds being examined occurred (mean values of 0.9 mg MBAS/L against 1.3 mg MBAS/L at the inlet, were reported). In the case of non-ionic surfactants, a considerable removal efficiency was found (higher than 99% on average). In this case, a significant removal increase was noticed, both with respect to the conventional activated sludge treatment and to the ozone treatment; the latter, as everyone knows, shows a lesser efficiency with respect to MBAS. Also in the case of non-ionic surfactants, the removal operated by the MF membranes can be deemed negligible (Chang et al., 2001; Marcucci et al., 2002), therefore we believe that the removal increment has to be ascribed to an actual increase in the biological removal. On the other hand, in the textile wastewater, most of non-ionic surfactants are made up of ethoxylate compounds (The Society of Dyers and Colourists, 1990), for which biodegradation possibility exists, according to different methods (please refer to Maki et al., 1994; Tidswell et al., 1996). Table 3 Data relevant to surfactant removal (values in mg/l) Inlet Permeate Secondary effluent Ozone outlet MBAS Average STD Max Min Non-ionic surfactants 118 Average STD Max Min

7 Conclusions The pilot-scale plant demonstrated that the MBR treatment makes it possible to obtain high purification efficiency of textile wastewater. Extremely satisfactory results were obtained both on conventional parameters such as COD, suspended solids, ammonium, and on compounds typical of this type of wastewaters such as dyes and surfactants. In the case of dyes and surfactants, removal efficiency similar or higher than that obtained with the Baciacavallo WWTP complete chain, were reached. These results appear to be extremely important, but only a very limited literature is available on them. As regards treatment applicability, it is advisable to specify that the system adopted provides for the use of plate and frame membranes with an external module. This system has high energy consumption and is suitable for small flow rates treatment (2,000 3,000 m 3 /d) with high pollutant concentrations. It would be therefore proper to experiment also alternate membrane typologies (for example, hollow fiber membranes) more suitable for high flow-rates treatment. Acknowledgements GIDA SpA technicians and researchers have contributed to this research. GIDA SpA have also provided financial support in the different activities, besides housing the pilot-scale plant. We would like to thank Air Liquide for making the oxygen supply system available and, particularly, for the collaboration provided in the research execution. We also thank Rhodia for placing the pilot plant at our disposal. References Chang, I.S., Chung, C.M. and Han, S.H. (2001). Treatment of oily wastewater by ultrafiltration and ozone. Desalination, 133, Germirli Babuna, F., Soyhan, B., Eremektar, G. and Orhon, D. (1999). Evaluation of treatability for two textile mill effluents. Wat. Sci. Tech., 40(1), Maki, H., Masuda, N., Fujiwara, Y., Ike, M. and Fujita, M. (1994). Degradation of alkylphenol ethoxylates by Pseumonas sp. Strain TR01. Appl. Environ. Microbiol., 60(7), Mamais, D., Jenkins, D. and Pitt, P. (1993). A Rapid Physical-Chemical Method for the Determination of Readily Biodegradable Soluble COD in Municipal Wastewater. Wat. Res., 27(1), Marcucci, M., Ciardelli, G., Matteucci, A., Ranieri, L. and Russo, M. (2002). Experimental campaigns on textile wastewater for reuse by means of different membrane processes. Desalination, 149, Orhon, D. (1998). Evaluation of industrial biological treatment design on the basis of process modelling. Wat. Sci. Tech., 38(4 5), 1 8. The Society of Dyers and Colourists (1990). Colorants and Auxiliaries, J. Shore (ed.), vol 2, 1st edn, Manchester, UK. Tidswell, E., Russell, N. and White, G. (1996). Ether-bond scission in the biodegradation of alcohol ethoxylate nonionic surfactants by Pseudomonas sp. strain SC25A. Microbiology, 142, Water Research Commission of South Africa (2000). Waste minimisation guide for the textile industry: a step towards cleaner production, WRC Report No TT 139/00, vol 1, prepared by Barclay, S. and Buckley, C., Water Research Commission of South Africa. 119