Efficacies of a locust bean gum polymer on the startup of a novel upflow anaerobic sludge blanket reactor treating municipal sewage

India being a developing country, is in the phase of rapid industrialization and urbanization. This is well evident from the agglomeration in urban population from 28 crores in 2001 to 37 crores in 2011 (taken from Census 2011). Though both are good on one hand with respect to our economic developments, it has indeed put a threat to the goal of sustainable development (Aishwaryalakshmi & Palanivelu 2022). Urbanization has imposed intense pressure on our water resources in two major tracks. Rise in water demand to meet the domestic requirements, and discharge of this used water on the receiving environment. To satisfy the growing demand, water for domestic and industrial needs is tapped from rivers, streams, wells, lakes and various groundwater resources. Of which 80% of the supplied water for domestic needs returns back as wastewater, depleting the water bodies (Sangamnere et al. 2023).

According to the reports of the Central Pollution Control Board (CPCB) (2021), there are nearly 1,631 sewage treatment plants (STPs) operating in different parts of our country with a total capacity of 36,668 million litres per day (MLD). Surprisingly the quantity of sewage generated by urban localities was estimated to be 72,368 MLD (nearly double the capacity of treatment plants). Though there are many treatment schemes which are strategically planned and are in place, yet there is always a large gap between the quantity of sewage generated and the capacity of treatment plants. Hence alarming issue faced by every developing/underdeveloped country would be the discharge of untreated or partially treated municipal sewage into water bodies (Henze & Ledin 2004). Having finite resources in hand, the only sound option to meet our demands would be the conservation of the existing aquatic ecosystem. Municipal sewage in other ways, can be considered as a resource to meet our non-potable requirements. By which the groundwater and surface water resources can be conserved (Mazhar et al. 2021).

With the rising concern for net-zero carbon emission, anaerobic treatment systems can be considered as the wise choice (Wang et al. 2005). For every kg of chemical oxygen demand (COD) anaerobically digested, nearly 0.35 m³ of biogas and 0.1 kg of biomass are obtained. On average, about 1 kWh (fossil energy) can be recovered from kg⁻¹ of COD destroyed (which in terms of CH₄ is 13.5 MJ). This energy-rich biogas (with 50–75% methane) can be used as an alternative to fossil fuels, thereby reducing global carbon dioxide (CO₂) emissions. Among the anaerobic systems, the granule-based UASB reactors have gained an extensive reach, because of their high treatment potential (Lettinga et al. 2001). Since biomass retention and liquid retention are uncoupled in a UASB reactor, they can maintain a high loading rate with longer solids retention time (SRT) even at short hydraulic retention times (HRTs) (Arivalagan & Stanislaus 2022). Hence, the reactor's volume can be considerably reduced, which further would cut down the capital cost, making it a cost-effective technology.

Generally, UASB reactors are coined as energy-economic systems, as they are apparent in forming their own sludge bed (Guo et al. 2022). Granules which come up with the classic aggregation of biomass are at the heart of this treatment scheme. These granules are featured with enhanced settleability, which thereby holds the microbes within the system for a longer span (Sethi et al. 2023). Besides, they are capable of providing an in-depth contact between the substrate and microbiota, which would further enhance the treatment (Tiwari et al. 2005).

However, the major shortfall of this technology is its lengthened startup phase (time taken for the formation of granular sludge bed). On average it takes nearly 3–8 months for any UASB reactor to achieve a successful startup, relatable to the slow growth of methanogens (Owusu-Agyeman et al. 2021). To accelerate the startup phase and to extend the application of these reactors, researchers have proposed a number of additives. However, most of the findings were limited to high-strength wastewater. Hence, this study focused on the idea of using locust bean gum (LBG), a natural gum-based polymer, as an additive to accelerate the startup of the UASB reactor treating municipal sewage. This water-based viscosifier (LBG) holds a good coagulating potential and exotic self-gelling property, which remained as the reason behind its selection. Also being a non-ionic polymer, its solutions were seldom influenced by pH, salts and heat treatment. Going for biopolymers would in turn benefit us with the ease of degradation – posing no harm to the ecosystem (Liang et al. 2019).

Hence, a study was taken up to evaluate the startup of a novel UASB reactor (designed to bear the resemblance of a clariflocculator, with a centric UASB unit and outer concentric downflow hanging sponge (DHS) module) in comparison with the conventional system. Effluent and sludge characteristics of both reactors were analyzed in detail to pin down the influence of additives used, in enhancing the startup of the UASB reactor treating municipal sewage.

Seed sludge for the study was collected from the anaerobic digester of an STP at Perungudi, Chennai, India. The inoculum was screened through a 0.15-mm sieve to remove the floating debris and large-sized inert fractions and then seeded to the reactors. Detailed characterization of the seed sludge is presented in Table 1.

Feed for the reactors was collected fresh at the start of each day, from the STP at Anna University, Chennai, India. To begin with, the wastewater collected after screening was characterized for its qualitative and quantitative parameters, following APHA (2012) standards, and these are tabulated in Table 2 (Water Environment Federation 2012). The substrate exhibited higher organics (COD 720–880 mg/L) and nutrient concentrations (TP 22–35 mg/L) than the city's municipal sewage, as the STP receives wastewater from the university's toilets, hostels, mess kitchen, cafeterias, laboratories, laundry units and staff quarters.

LBG is a creamy white powder, produced by grinding the seed endosperm of the carob tree (Ceratonia siliqua). In general, it is predominated by the complex carbohydrate polymers of galactose and mannose in different proportions (say 1:3.1–1:3.9). Also, the average molecular weight of locust galactomannan varies typically in the range of 0.3–2.0 million. This galactomannan is known to exhibit an extended ribbon-like pattern in its solid state and a semi-flexibile coil framework in its liquid state (Barak & Mudgil 2014). Besides its polysaccharidic structure, the viscosity induced by the polymer facilitates the aggregation of microbiota through hydrogen bonding (Cheng et al. 2020).

(1) Sludge bed zone (0.10 m Ø) – rising to a depth of 0.45 m from the inlet port. (2) Filter media zone (0.10 m Ø) – comprised of Polyvinyl alcohol (PVA) gel media (procured from Kuraray Co., Ltd – Tokyo, Japan), floating against a screen at 0.70 m height. (3) Enhanced gas liquid solid separator (GLSS) of 0.25 m height – the lower half of GLSS was designed as an inclined entity with a slope angle of 60°, whereas the upper half resembled a tubular unit of 0.25 m Ø. To capture the biogas generated, the top lid of the GLSS facility was anchored with a canopy of 0.23 m Ø. The captured biogas was quantified by the water displacement method, using a Mariotte bottle (Mamun & Torii 2015). Effluent from the GLSS was uniformly distributed to the outer aerobic compartment by a perforated plate. DHS module comprising of 6th generation sponge cylinders (supplied by courtesy of AKSHAT enterprises – India) was incorporated as the polishing component, to remove the nutrients and remaining carbonaceous fractions. Finally, the treated water was collected from the effluent ports positioned at the bottom.

The photographic setup and line sketch of the conventional UASB reactor (RC) fabricated for this study is represented in Figure 1(c) and 1(d). The plexiglass-made reactor of 8 mm thickness provided a working volume of 6 L (similar to that of RH). A cylindrical column of 0.15 m (Ø) extending to a depth of 0.45 m stood as the functional unit of RC. Feed was supplied to the reactor from the inlet port at the bottom and the required flow rate was maintained with the peristaltic pump, procured from Susethil Engineering, Chennai.

The sludge bed zone occupied a depth of 0.15 m from the inlet port, above which was the clarifier zone (observed at the middle third of the reactor). A conventional GLSS of (0.15 m height) was stationed at the top end of the reactor. It comprised of an inverted cone (0.12 m Ø) affixed with the lid of the reactor – biogas generated from the system was collected through this provision. At this juncture, treated water gets collected from the effluent port with a water seal facility.

Influent and the effluent samples along with those drawn from various ports across the reactors were analyzed for pH, alkalinity, COD, volatile suspended solids (VSS), and volatile fatty acid (VFA), on every alternate day. Sludge samples drawn from the bottom of both the reactors were analyzed for mixed liquor suspended solids (MLSS), mixed liquor volatile suspended solids (MLVSS) and sludge volume index (SVI) on every other day. All the analysis were carried out in accordance with the ‘Standard methods for examination of water and wastewater’ (APHA Standards 2012).

A digital pH meter procured from Vani International, India was used for the determination of pH. The COD of the samples were analyzed following the closed reflux colorimetric method. VFA was quantified by titration method following (Dilallo & Albertson 1961). The first 25 mL of the sample was taken and its initial pH was noted, it was then titrated with 0.1 N of H₂SO₄ until a pH of 4.3 was observed. The sample's pH was further lowered to a range of 3.3–3.5 and boiled to remove excess CO₂. It was then cooled to room temperature and back-titrated to a pH of 7 using 0.05 N of NaOH. Based on the volume consumed during both titrations, VFA concentration was estimated.

Since extracellular polymer (ECP) content comprising (polysaccharides and protein) plays a dominant role in the granulation process, it was also quantified. Extraction of ECP from the sludge samples was done by cooling extraction method (Kuba et al. 1992). Two milliliters of sludge sample was centrifuged to remove the supernatant and added with 10 mL of 0.85% NaCl and 60 μL of formalin. This mixed liquor was ultra-sonicated for 300 s while being in an ice pack, for the extraction of ECP content. Later the mixture was centrifuged at 12,000 rpm for 30 min and the supernatant was alone separated and analyzed for polysaccharide and protein. The sulfuric acid-anthrone method was adopted for the estimation of polysaccharide and the Lowry Folin method helped with protein quantification (Lowry et al. 1951).

Serum bottle technique was taken up for the estimation of maximum specific methanogenic activity (SMA), using sodium acetate as the sole substrate at 35± 1 °C under anaerobic conditions (Valcke & Verstraete 1983). With the passage of time, the concentration of substrate from the serum bottle was assessed. The maximum slope of the concentration curve plotted denoted the activity of methanogenic bacteria.

Periodic visual examination of the granules was done and their surface morphology was studied by scanning electron microscopy (SEM). The sludge samples were prepared for analysis by fixing in a 2.5% glutaraldehyde solution, dehydrating them in graded water–ethanol solutions (30, 50, 70, 80, 90, 96 and 100%), followed by critical point drying and finally sputter coating with gold. SEM images were then taken with the support of the JEOL JSM-500LV microscope (Arivalagan & Stanislaus 2022).

Fourier transform infrared spectroscopy (FTIR) technique was adopted to detect the functional groups present in the granules and to get more insights on the mechanism of their formation. The sample for analysis was prepared by subsequent freezing and drying. Potassium bromide (KBr) pellets for analysis were prepared by mixing 1 mg of powdered sample with 100 mg of spectrometry grade KBr under vacuum. FTIR spectra were then taken on the prepared KBr pellets using a Bruker VERTEX-70 IR infrared spectrometer.

The LBG polymer used for this study was procured from ‘Chem pro solutions’ at an average cost of Rs. 500/kg. RH reactor was started by blending this polymer additive at a concentration of 20 mg/g TSS – from the comparative study of (Lakshmi & Raj 2022), with the seed sludge and the prepared inoculum was allowed to occupy one-third of the reactor's volume. Whereas RC started off with the collected seed sludge alone (with no additives), to compare the effectiveness of LBG polymer in accelerating the startup phase. The seed sludge was left undisturbed in both reactors for 3 days – to help microbes acclimatize to the newer environment. As suggested by (Barber & Stuckey 1999), both reactors were then fed with domestic sewage at a very low loading rate of 0.87 kg COD/m³/d, maintaining an HRT of 24 h. The performance of the reactors during startup was evaluated, by monitoring both the effluent and sludge characteristics.

Though the acclimatization period of both reactors was almost similar, RC witnessed a longer span of growth phase (18–85 days) with a gradual rise in removal efficiencies. Hence the reactor reported the highest removal of 83% on the 85th day, which was steadily maintained for the next 10 days confirming a successful startup. This difference in the startup of both reactors confirms the influence of the LBG polymer in accelerating granulation. The polymer by its long chain length has aided in microbial bridging and its viscous nature in turn promoted the contact between microbes (Barak & Mudgil 2014). This proximity further enhanced mass transfer between syntrophic groups in the granules, which has indeed resulted in the improved removal of organic matter (Liang et al. 2019). Hence the polymer was also effective in reducing the startup time by nearly half than the control system. (Kalogo et al. 2001) took 22 weeks for the startup of a self-inoculated UASB reactor, with the support of the water extract of Moringa oleifera seeds (WEMOS) additive. Similarly, chitosan as an additive (Tiwari et al. 2005) witnessed startup in 65 days, with the largest granules of mean size 0.15 mm.

Figure 4(b) presents a comparison of effluent alkalinity and VFA during the startup of reactors RH and RC. During the acclimatization phase, both reactors were observed with a rise in VFA concentration from (20–40 mg/L) as the rate of consumption of VFA by microbes was less – relatable to the observations of (Wang et al. 2018). Thereby pile up of VFA concentration has resulted in reduced alkalinity (272–305 mg/L as CaCO₃) and pH (6.6–6.8) during this stage as evident from Figure 4(b). With the onset of methanogenesis, VFA concentration dropped steadily reaching (8.5–10 mg/L) for RH and 17 mg/L for RC, toward the end of the growth phase. This period also visualized a consequent increase in alkalinity (320–390 mg/L as CaCO₃) and pH (6.9–7.4) – similar to the results of (Kalogo et al. 2001; Zhang et. al. 2009). VFA/alkalinity ratio observed during the startup phase of RH and RC reactors was in the range of (0.03–0.13), which is less than 0.5 – an indication of proper anaerobic functioning, relatable to the observations of (Loganath & Mazumder 2018). (Ravichandran & Balaji 2020) in their study on a hybrid upflow anaerobic sludge blanket reactor (HUASB) reactor, observed VFA/Alkalinity ratios in the range of 0.13–0.381.

Sludge characteristics of both reactors were monitored, to connect the role of polymer additive with the acceleration of startup. A record of the microbial population (VSS and TSS concentrations), their settleability (SVI), and activity (SMA) at periodic intervals was established.

This was further strengthened by ECP characterization of sludge samples, as shown in Figure 6(b). On completion of startup both reactors were characterized with ECP concentrations in the range of 0.30–0.32 g/g VSS. Corroborating that early startup was induced by the polymer additive, which was later strengthened by ECP secretions (Show et al. 2004). Similarly (Guo et al. 2022) in their study using bamboo charcoal witnessed ECP concentrations in the range of 0.1–0.26 g/g VSS and (Guo & Kang 2020) also observed similar protein and polysaccharides concentration of 0.03 and 0.12 g/g VSS. Whereas (Liang et al. 2020) quantified lower ECP concentrations in the range of (20–27 mg/g of VSS).

A heterogeneous microbial community dominated by cocci (Methanosarcina), bacilli (Methanothrix), and filamentous bacterium was spotted in the samples of both reactors – as observed from Figure 7(d) and 7(e). The heterogeneity witnessed here was due to the complex nature of real wastewater (rather than synthetic) treated. According to De-Kreuk & van Loosdrecht (2006), heterogeneity is a distinct feature of granules which thrive on real wastewater. Also, Tay & Yan (1996) have made an interesting remark that the substrate used can influence the microbes contained in a granule. This was cited with an example – filamentous bacteria were found abundant in glucose-concentrated substrates.

Since the major bands of carboxylic, amide, and carbohydrate functional groups fell in the region of 1,800–900 cm⁻¹, it was dealt with in detail. Two peaks observed at 1,737 and 1,367 cm⁻¹ were attributed to the C = O stretching vibration of three-turn helix and β-sheets in secondary protein structure, which has favored bio-flocculation. The presence of dissociated carboxyl groups could support the coagulation process by forming ion bridges, followed by the binding of divalent metal ions (Awang & Aziz 2012). In addition, the shoulder peak spotted at 1,520 cm⁻¹ was due to the ring vibration in phenols of tyrosine side-chains, which confirmed tyrosine protein as a structural component of mature granular sludge (Dong et al. 2017).

The peaks spotted at 975 and 982 cm⁻¹ indicated weak vibrations from phosphodiester bonds hinting at a lesser amount of extracellular nucleic acids. This can be taken as a positive sign in terms of microbial health. Hence, lower extracellular nucleic acid content confirmed higher biological activity of sludge, which was also apparent from the reactor's high treatment potential.

This work explored the potential of a natural gum-based polymer – LBG – in enhancing the startup of UASB reactors handling municipal sewage. A comparative study was taken forward to ascertain its competence and the following conclusions were drawn.

RH reactor started with an LBG polymer and achieved a successful startup in 40 days – with the highest COD removal of around 95%, in accordance with the effluent concentrations of (37–50 mg/L). Whereas RC took nearly 85 days to achieve the highest COD removal of 83%. The accelerated startup of RH sheds light on the potential of the LBG polymer in immobilizing biomass within the sludge bed and the eminence of the reactor's modified version – which has reflected on its improved removal. Relatively the biomass concentration of RH was observed at the highest of 20–22.1 g/L on the 45th day, corresponding to an SVI of 26 mL/g (the sign of a dense sludge bed). Besides RC took nearly 85 days to achieve the highest biomass concentration of 21.52 g/L, with respect to an SVI of 30.5 mL/g. SMA and morphological study of granules in RH further confirmed the presence of active methanogens, hence the shorter granulation period had no compromise with the microflora and their activity. Since the chosen biopolymer holds the ease of degradation, they do not cause any potential harm to mother earth. Hence, in a nutshell the polymer has proved its competence and can be considered as a sustainable choice for enhancing anaerobic granulation. The usage of this additive is yet to be investigated for several high-strength industrial wastewaters, which would fetch them a wider application.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.