A pilot-scale study on PVA gel beads based integrated fixed film activated sludge (IFAS) plant for municipal wastewater treatment

Secondary treatment of municipal wastewater is usually accomplished by biological processes, which can be classified as suspended and attached growth processes (Singh et al. 2014, 2015a, b). Among the suspended growth processes, the activated sludge (AS) process is one of the most commonly used methods for biological treatment of municipal wastewater (Malmqvist et al. 2004; Singh et al. 2015a, b). Nowadays, most of the AS plants are suffering from frequent problems, most often due to unwanted biosolids-liquid separation (Jenkins et al. 1993). The low solids separation efficiency can lead to increased concentration of (1) organisms (measured in terms of biological oxygen demand (BOD)/chemical oxygen demand (COD)), (2) solids (in terms of total suspended solids (TSS), volatile suspended solids (VSS), etc.) and (3) nutrients (in terms of N and P bearing compounds) in the treated effluent; something which can then impact negatively on wastewater recyclability as well as the receiving environment (Akker et al. 2010). This problem may be minimized by larger bioreactor volumes, which can be very expensive and sometimes impossible if space is limited (Almomani et al. 2014). Therefore, there is a need for compact hybrid systems that can increase capacity within existing volumes.

In last two decades, there has been an increasing interest in development of integrated fixed film activated sludge (IFAS, also known as hybrid biological reactor) systems for municipal as well as industrial wastewater treatment. These systems combine features of suspended and attached growth processes by incorporating specially designed biomass carriers, on which the biomass attach and populate, in bioreactors. This addition of biomass carrier increases the biomass inventory and subsequently the treatment capacity of the reactor (Ødegaard et al. 1994; Randall & Sen 1996; Rusten et al. 2006; Wang et al. 2006; Rouse et al. 2007; Seetha et al. 2010; Li et al. 2012; AnoxKaldnes 2014; Chan et al. 2014). Based on arrangement of carrier, IFAS systems are categorized into two types: moving bed biofilm reactors (MBBRs), in which carrier moves freely inside the reactor, and fixed media activated sludge systems, in which carrier is fixed inside the reactor (Ødegaard et al. 1994; Randall & Sen 1996; AnoxKaldnes 2014).

Among these two modules of IFAS systems, MBBRs have gained much attention in developed countries, but very limited studies are available in developing countries such as India (Anterrieu et al. 2014; Jabari et al. 2014; Piculell et al. 2014; Zhang et al. 2014). These systems require low construction cost, minimal space requirement, simple operation combined with effective removal of organic and inorganic pollutants (Almomani et al. 2014; Yang et al. 2014), low sludge production, high biomass concentration, long sludge residence time (SRT), lower head loss and no clogging problems (Andreottola et al. 2000a, b; Ødegaard 2006; Rutt et al. 2006; Kim et al. 2010; Rosso et al. 2011; Chen et al. 2014). Due to all these specific features, various biomass carrier based MBBRs have been applied for the treatment of wastewaters in laboratory-scale, pilot-scale, and full-scale systems across the world (Sen et al. 1994; Ochoa et al. 2002; Trapani et al. 2010; Bassin et al. 2012). The carriers employed in MBBRs are designed to provide a large protected surface area and the key materials determining the efficiency and performance of treating wastewater in MBBRs (Larrea et al. 2007; Levstek & Plazl 2009). These biomass carriers are mainly grouped into two categories: porous and non-porous carriers. The non-porous carriers are usually made from polyethylene or polypropylene and have high durability and stability, and are widely used in the wastewater treatment plants and in the laboratory. However, they have some drawbacks of low specific surface area (∼500 m²/m³), leading to slow start-up and easy detachment of biofilm due to the smooth surface. On the contrary, the porous carriers can escape from those drawbacks due to the excellent structure, which can not only trap and intercept the biomass efficiently as well as shorten the start-up period, but also promote biofilm accumulation by providing a large surface area (Falletti & Conte 2007; Sabzali et al. 2013; Chen et al. 2014). The applications of these types of carriers are still in their infancy due to lack of knowledge about their potential efficiency and application in conventional wastewater treatment systems.

Recently, polyvinyl alcohol (PVA™) gel beads (a porous biomass carrier), which have been shown to be effective for cultivation and retention of slowly growing bacteria, are attracting attention from many researchers. Successful implementation of MBBRs incorporating PVA gel beads as biomass carrier has also been reported for the treatment of municipal and industrial wastewater (Rouse et al. 2005; Rouse et al. 2007; Zhang et al. 2007; Levstek & Plazl 2009). These beads are typically used at volumetric packing ratios of only 5 to 15% versus much higher ratios of 50 to 70% common to other non-porous carriers, thus allowing more hydraulic loading to these systems (Levstek et al. 2010).

Among the various treatment configurations of MBBRs, recently two processes named as Hybas™ and BAS™ (AnoxKaldnes 2014), a combination of MBBR and AS processes, have shown significant potential in wastewater treatment (Chan et al. 2014). A similar configuration named the bio-film activated sludge (BAS) process, in which MBBR is followed by the AS process, has been tested successfully for food industry, textile, paper and pulp, petrochemical and dairy industry wastewaters, although its potential for municipal wastewater treatment is still to be investigated (Dalentoft & Thulin 1997; Malmqvist et al. 2004, 2007; Werker et al. 2004; Sivard et al. 2007; Haandel & Lubbe 2012). These systems are able to guarantee high COD removal performance at sludge age equal to the AS system, or similar performance at lower sludge age. The performance of a BAS process operated at 6–10 days of apparent sludge age and a F/M ratio of 0.2–0.4 kg COD. kg⁻¹. VSS.d⁻¹ is comparable to that of a conventional activated sludge system with a sludge age of 20 days and a F/M ratio of 0.2–0.25 kg COD.kg⁻¹. VSS.d⁻¹ or less. The treatment volume required by the BAS process is also 50–70% less than that of the AS process (Haandel & Lubbe 2012).

In this context, the feasibility of a PVA gel beads based IFAS (in BAS mode) plant for municipal wastewater treatment was addressed in this study to achieve effluent quality below the discharge standards of Punjab Pollution Control Board (PPCB), India (Table 1). In addition to this, the microbial community of this system was also investigated by culture dependent and independent techniques (Bhatia et al. 2013).

A pilot-scale IFAS plant was installed and operated in BAS mode at the sewage treatment plant (STP) of Jalandhar, Punjab, India. The pilot-scale plant consisted of two reactors (R1 & R2), each having 10 L capacity, in series followed by a secondary clarifier of 5 L capacity. The sludge was separated in a clarifier (5 L) and returned to the R2 reactor. Both reactors had the provision for excess sludge purging. Description of each reactor is given below:

R1 reactor: This reactor was filled with biomass carriers with 10% packing ratio and operated as MBBR. Influent wastewater was pumped to this reactor and its outlet directed to R2 reactor. Sieve arrangements were adopted to retain the carriers inside the R1 reactor. This reactor was operated with high dissolved oxygen to keep the carriers in suspension.

R2 reactor: This reactor was operated simply in activated sludge process mode, and settled activated sludge from clarifier was returned to R2 to maintain biomass in appropriate proportion. The excess sludge was disposed of from the clarifier daily.

PVA™ gel beads, a trademark biomass carrier from Kuraray Co., Tokyo, Japan, were used in the present study. They are hydrophilic in nature and have a very porous structure with only 10% solids and a continuum of passages of 10 to 20 μm in diameter tunneling throughout each bead. This carrier allows the growth of aerobic bacteria on the outer surface of the carrier and anaerobic growth at the core, with occupying capacity of 1 billion microbes per bead (Ramteke et al. 2015; Kuraray Aqua Ltd 2015). Typical characteristics of these carriers are shown in Table 2.

Reactor performance was monitored by analysing the water quality parameters of influent and effluent samples of the pilot plant. Spot samples (daily basis in phases 1 and 2; alternate days in phase 3) were collected and analysed for BOD, COD, TSS, VSS, ammonia nitrogen (NH₃-N), nitrate nitrogen (NO₃-N), and ortho and total phosphorus (OP and TP). Samples for determination of soluble components were passed through Whatman 42 (0.45 μm) filter paper prior to analyses. Activated sludge from R2 tank was also characterized by mixed liquor suspended solids (MLSS), mixed liquor volatile suspended solids (MLVSS) and sludge volume index (SVI). All analyses of spot samples were conducted in accordance with Standard Methods (APHA 2005). Other operational parameters, like dissolved oxygen (D.O.), pH, temperature, flow rate and hydraulic retention time (HRT), were also monitored at the pilot plant site. The pH and D.O. were measured by a table pH meter (Cyberscan 510 digital, Thermo Scientific, Mumbai, India) and a portable D.O. meter (HACH- Model OX-2P, Hach, Bangalore, India), respectively.

Total and faecal coliforms (TC and FC) were measured according to Standard Methods (APHA 2005). The pour plate technique and serial dilution method was used to enumerate the bacterial communities in the pilot plant. Total heterotrophic bacterial (HPC) count, fecal Staphylococcus and Escherichia coli were determined as described in Bhatia et al. (2013).

Attached and suspended sludge samples were collected from the R1 reactor of the pilot plant. Culture independent methods using polymerase chain reaction (PCR) amplification of specific 16S rRNA sequences (16S ribosomal RNA) were used for identification of dominating bacteria in suspended and attached biomass (Bhatia et al. 2013). The experimental analysis was performed in the following steps. The total genomic DNA was extracted from samples with the bead-beating method, using a DNA extraction kit (Hi PurA™ Soil DNA kit, Hi Media, Mumbai, India) as per manufacturer's instructions. The universal primer pair 1492 r (5′-TAC CTT GTT ACG ACT T-3′) and 27f (5′-AGA GTT TGA TCC TGG CTC AG-3′) was used to amplify the 16S rRNA gene. The PCR reaction was done using 2 μM of primers, 50 ng of genomic DNA, 200 μM of each dNTP, 1 × PCR buffer, 2 mM MgCl₂, and 2.5 units of Taq DNA polymerase (Biochem Biotech products, Sigma Aldrich, Bangalore, India). The PCR reaction was carried out at 94 °C for 2 min, followed by 35 cycles at 94 °C for 1 min, then at 48 °C for 1 min and 72 °C for 1.5 min. The final elongation step was carried out at 72 °C for 10 min (Bhatia et al. 2013). Thereafter DNA sequencing was done (BioAxis DNA Research Center (P) Ltd, Hyderabad, India) and sequence data from 16S rRNA gene fragments were analysed by a Genetic analyser (ABI model 3100, Applied Biosystems, New Delhi, India) and compared using the biocomputing tools provided online by the National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov). The basic local alignment search tool (BLAST) program (Altschul et al. 1990) was used for sequence similarity analysis.

Characterization of ciliate species was done by taking 25 μL sub-samples of mixed liquor with an automatic micropipette, and a minimum of four replicates of this volume were counted each time. For the image acquisition, a drop of sludge was deposited on a slide and carefully covered with a cover slip. Image acquisition of the flocs on the slide was carried by phase-contrast illumination technique (magnification ×100) using a photonic microscope (Olympus Medical Systems India Private Limited, Gurgaon, India) and keeping the illumination constant for all samples (Martín-Cereceda et al. 2001).

At start-up, the pilot plant was inoculated with 5 L of activated sludge from the SBR plant (5 day SRT) located in Jalandhar. The pilot plant was operated for 4 months, in three consecutive phases with varying D.O., influent flow (Q) and external recycle ratio (ERR). The operational parameters in each phase are shown in Table 3.

At optimum conditions (phase 3), half of the inflow and effluent from MBBR were fed to the AS tank as shown in Figure 1. Typical composition of influent fed to plant is shown in Table 4. It was necessary to maintain appropriate food (BOD) to microorganism (MLSS) ratio, i.e. F/M (0.2–0.5) in the R2 reactor, so that it could behave like a conventional activated sludge process (Metcalf et al. 1991).

In the initial phase of process optimization, sewage temperature was high, nearly around 30–35 °C. This temperature range was found to be detrimental to microbial species and the sludge settleability, and resulted in floating sludge in the clarifier (Malmqvist et al. 2004). Another reason behind this unwanted phenomenon may be denitrification at the bottom of the settling tank, which is very common in AS based systems (Wanner 1994). Beside this, the recycling of an inappropriate amount of active biomass could be a possible reason during the process (Metcalf et al. 1991; Qasim 1998). However, the process was optimized by maintaining suitable operational conditions (ERR and D.O. were maintained as shown in the third phase of optimization), and this improved the settleability of the activated sludge. After the optimization, the process has shown stable results for long periods. Operating parameters of each reactor of the pilot in third phase (optimum conditions) are shown in Table 5.

During the study period, high variation was also observed in influent quality, i.e. COD ranging from 48 to 1,118 mg/L and BOD ranging from 28 to 456 mg/L. The average concentrations of BOD and COD in influent were also quite high, as shown in Table 3. But the performance of the reactor was observed to be stable at optimum state. This indicates that the system is able to bear the organic shock load and provided consistent values, indicating good quality of effluent in terms of BOD and COD. The results obtained in this study were in good agreement with previous studies (Andreottola et al. 2000a, b; Javid et al. 2013; Van Haandel & Van der Lubbe 2015).

Sludge from the AS tank of the pilot plant was characterized in terms of MLSS, MLVSS and SVI. After optimization, the sludge characteristics improved slowly, and SVI reached up to 60 mL/g on the 110th day. Typical values of SVI varied in the range of 25–72 mL/g at stable conditions. These values of SVI represent good settling in the current treatment scheme. For a pure suspended growth system, the sludge yield coefficient varies in the range of 0.4–0.6 g per g of COD removed (Metcalf et al. 1991), while in the present study sludge production was calculated as 0.18 kg VSS/kg COD removed at optimum condition. This indicates that the sludge production was much lower in comparison with other suspended growth systems.

Microbiological experiments were performed for pilot influent and effluent at optimum conditions (third phase in Table 3). The results demonstrated that the removal percentage for TC, FC, HPC, E. coli, and Staphylococcus was ∼96%, 99%, 99%, 99% and 85%, respectively.

To explore the microbial consortium in PVA gel, 16S rRNA gene based identification (molecular method) was done. The 16S rRNA gene sequences of the three most dominant distinct bacteria were detected to be most similar to gene sequences of E. coli, Pseudomonas sp. and Nitrosomonas communis (98, 97, and 98% similarity), respectively. The presence of rod shaped Gram-negative E. coli bacteria confirms the anaerobic zones in biofilm of carrier biomass. All Pseudomonas sp. are rod shaped Gram-negative and are supposed to be responsible for biofilm formation, while Nitrosomonas communis are mainly responsible for high nitrification rates in the current treatment scheme (Siripong & Rittmann 2007; Qin et al. 2008; Wang et al. 2010).

Moreover, ciliate species (protozoa) are one of the most common components in anthropogenic systems (biological wastewater treatment plants) and play an important role in wastewater treatment processes. These species are responsible for improving the quality of the effluent, maintaining the density of dispersed bacterial populations by predation. Studies of the relationships between protozoa and physicochemical and operational parameters have also revealed that the species structure of these communities is an indicator of plant efficiency (Madoni 2011). The indicator species of the representative group of ciliated protozoa growing in the activated sludge of the IFAS plant are shown in Table 6. The most dominant species of ciliates identified in the IFAS plant were of the attached type, as reported earlier (Madoni 2011).

The results of the present study demonstrated the steady state efficiency of a PVA gel beads based IFAS plant with respect to removal of organics and nutrients (N and P forms) from a real municipal wastewater, in spite of significant fluctuations in influent organic pollution load. The following insights were drawn from the present study:

• PVA gel beads were found to be a promising carrier which allows up-gradation of existing wastewater treatment plants with small packing ratio in comparison with other available biomass carriers, thus allowing more hydraulic loadings on treatment systems.

• The pilot-scale PVA gel beads based IFAS plant was found to be effective in municipal wastewater treatment. The removal rates of COD, BOD, TSS, and NH₃-N were 91%, 92%, 92.3% and 90.3%, respectively, which meets the discharge standards of PPCB, India. However, effluent discharge levels of total P could not meet the requirements of PPCB.

• SVI values obtained from the present study revealed that the sludge has good settling properties, having SVI in the range of 25–72 mL/g. The production of sludge has been determined as 0.18 kg VSS/kg COD removed, which was low as compared with stoichiometric values. Further research is required in this direction to reveal the facts behind low sludge production in these kinds of systems.

• Microbial examinations of PVA gel and activated sludge using 16S rRNA gene analysis showed richness of E. coli, Pseudomonas sp. and Nitrosomonas communis in PVA gel and Acinetobacter sp. in suspended biomass, respectively. Also, microscopic observation of activated sludge showed that Vorticella were the most dominant protozoa in the IFAS plant.

• The results obtained in the present study may aid in the design of full-scale processes. In addition, these innovative processes can be applied to existing plants to enhance the treatment efficiency without major new construction.

The authors are thankful to the Kuraray India Pvt. Limited, India, for financial and technical support to this study.