Anaerobic codigestion of urban solid waste fresh leachate and domestic wastewaters: Biogas production potential and kinetic

2.1. Landdfill Leachate and Domestic Wastewater Sampling

The leachate used was directly collected from the wastes compactor vehicles before waste dumping at the sanitary landfill of Béni Mellal city, Morocco. The landfill site covers approximately 200,000 m² and receives about 160 tons of municipal solid wastes per day with an amount of daily fresh leachates production estimated to 20–60 m³, depending on the season. The full characteristics of this leachate are reported on Table 1. It showed typical composition of the young leachate used with relatively high level of chemical oxygen demand (COD) (42.5 g/L), total Kjeldahl nitrogen (TKN) (2,960 mg/L), ammonia (765 mg/L), volatile fatty acid (VFA) (14.6 g/L) and salts content (24.3 ms/cm). Wastewater used was collected at the municipal wastewater plant treatment; its chemical oxygen demand is low and typical of domestic wastewaters.

Table 1

Leachates Characteristics

2.2. Codigestion Ratios Preparation and Batch Experiments

Five volumetric ratios of leachate/wastewaters (L/W) were prepared for batch anerobic determination of biochemical methane potential: 100%L (R100), 75%L/25%W (R75), 50%L/50%W (R50), 25%L/75%W (R25) and 100%W (R00). The characteristics of each mixture are summarized in (Table 2).

Table 2

Characteristics of Used Leachates/Wastewater Ratios

The batch tests were conducted using standard reactor with working volume of 1.3 liter for a total volume of 1.5 liter. Each ratio was inoculated by 100 mL of granular anaerobic sludge collected from the municipal wastewater anaerobic reactor and then adjusted to pH 7.5 using KOH 0.1N. In order to simulate wastewaters treatment plant step, all the ratios were first air stripped for 48 h before anaerobic incubation. The anaerobic condition was assured through N₂ bubbling and then the reactors were sealed and kept at 35 ± 2°C using heated lab aquarium. The reactors were retired for 5 min manual agitation once a day and the biogas produced was measured every 3 to 4 d by the water displacement method. The incubation was carried out for a period of 90 d until no biogas was detected.

2.3. Kinetic Modeling

The cumulative biogas production kinetics obtained from anaerobic digestion of all batch reactors was assessed by fitting the experimental cumulative biogas data to the exponential rise to maximum and modified Gompertz kinetic model as described in literature [31, 33]. The corresponding equations used are as follows:

Exponential rise to maximum:

Modified Gompertz equation:

Where:

The nonlinear regression and analysis of experimental data was performed on the basis of least squares algorithm using MATLAB2015. The obtained coefficient of determination (R²) was used to evaluate the performance of the two models. For each set of data, R² was obtained by calculating the following ratio (3)

Where, xi and yi are the observed data and ū and ў are the calculated averages. The calculation of the correlation coefficient thus makes it possible to obtain an estimate of correlation degree between the experimental variables.

2.4. Analytical Methods

COD in the influent and effluent samples after each treatment was determined by the standard closed reflux colorimetric method [34]. The pH was monitored using a pH meter (HANNA HI2300), ammonia nitrogen concentration was measured using the Kjeldhal method and volatile solid (VS) were carried out following the procedures outlined in standard methods [34]. VFAs concentration was determined using the Volatile Acid Alkalinity Method described in Operations Manual Anaerobic Sludge Digestion [35].

3.1. The Effect of Air Stripping Pretreatment

The effluent ammonium and VFA concentrations before and after air stripping pretreatment are reported in (Table 3). Ammonium removal varied from 62 to 75% reducing the initial concentration to 265, 200, 186, 41 and 3.7 mg/L, respectively for R100, R75, R50, R25 and R00. Similar result were reported by others authors for which air stripping was operated at alkaline pH causing highly significant loss of ammonium by ammonia volatilization [36–39]. In our conditions the reduction of ammonium concentration during air stripping could be attributed much more to nitrification as the working pH was less favorable to ammonia loss by volatilization. Air stripping had also positive effect on VFA content with percent reduction that varied from 8% to 42% and this was attributed to VFA volatilization and biodegradation. Final values were 9.4 g/L, 7.4 g/L, 5.6 g/L, 2.1 g/L and 0.11 g/L, respectively for R100, R75, R50, R25 and R00. We expected an increase of VFA as the 48 h of air stripping is quite sufficient to stimulate microbial activity and leachate decomposition and hydrolysis, but it seems that volatilization of short-chain organic and odorous acids was predominant during this step.

Table 3

Ammonium and VFA Removal Rate by Air Stripping Pretreatment

3.2. Effect of Codigestion on Cumulative Biogas Production and Kenetic

Biogas production over the 90 d test is illustrated in Fig. 1. No significant biogas was detected for reactor R00 working with only wastewater and this was attributed to its low COD concentration either before or after air stripping. The lowest biogas yield was also recorded in case of leachate monodigestion (R100). For other ratios, biogas values and kinetic showed that codigestion had a positive effect and that leachate dilution by wastewater reduce the star-up period and improves biogas yield. Comparison between reactors showed that the highest biogas yield was achieved at R50 followed by R25. Similar results were obtained by others authors [40–42] regarding the positive effect of wastewater dilution. By diluting leachate by synthetic wastewater, optimal anaerobic digestion was also achieved at volumetric leachate ratio under 30% with maximum COD removal at 25% ratio [43].

Fig. 1

Biogas production rates (a) and cumulative biogas production (b).

Many parameters could be responsible for yield decrease at higher concentrations of leachates. Taking account the used leachate composition (Table 1) and as ammonium was at values favorable for methanogenises as documented by literature [44], the observed effects could be attributed to the overall initial composition of leachate and particularly the initial leachate COD and VFAs content that seems to be the most limiting factors in this case. Thus, even though high concentration of both organic COD loading and VFA was reported in literature for sludge anae robic digestion [45] and food waste leachate and sewage sludge mixture [46]. For VFA, it has been reported that toxic threshold depend on the kind of VFA involved and that toxicity can be enhanced significantly by synergism of the individual long-chain fatty acid [44, 47]. In recent work based on food leachate monodigestion and codigestion, the concentration of VFA was in good correlation with the performance of the reactors with values under 4 g/L as optimal concentrations [27]. For the present work, optimal codigestion condition where achieved within initial COD and VFA content range of 11.7 to 32.3 gO₂/L and 2.1 to 7.4 g/L, respectively as illustrated in Fig. 2. In terms of biogas yield this corresponded to a range of 350 to 438 mL/g VS. Biogas yield values closes to these finding were reported by recent works for both food wastes and leachate codigestion with sewage sludge [6, 46]. Results similar to our finding were also reported by [48] for which biogas yield varied from 233 to 344 mL CH₄/g VS for leachate wastewater ratio from 1:9 to 5:5 with maximum yield at 3:7 ratio.

Fig. 2

Cumulative biogas production versus initial COD and VFA concentration.

Results of modeling biogas production are reported in Fig. 3 and Table 4. Experimental data of cumulative biogas were best fitted with modified Gompertz model for which R² was around 0.999.

Fig. 3

Cumulative biogas modified Gompertz (a) and Exponential rise to maximum plot (b) where experimental data are given in symbols and simulation plots in lines.

Table 4

Key Kinetics Parameters According to Modified Gompertz Model

Exploration of modified Gompertz curves showed that leachate dilution improves the biogas potential and reduces the lag phase period of methanogenises from 25 to 6 d (Table 4).

According to the results obtained, the optimal anaerobic co-digestion conditions both in terms of maximal biogas potential, start-up period and maximum daily biogas production rate, could be achieved around leachate volumetric ratios from 25% to 75%. The maximum was recorded at 50%, even though decreasing ratio to 25% had postive effect on time needed to start methanisation and the hydraulic retention time needed to produce 90% of the BMP while increasing ratio over 50% had negative effect on both parameters. The maximum daily biogas production was around 16 mL/g VS for both 25% and 50% ratio but the HRT needed to produce 90% of the BMP was also shorter for the 25% volumetric ratio (31 d versus 41.5 d). Thus, although maximum BMP was recorded at 50% dillution it semms that optimal overall performance and stability of the codigestion could be achieved with more leachate dillution. These finding agree and are close to that of [43] using landfill leachate dilution with synthetic wastewater in the range of 5 to 75% (v/v) for which higher COD removal of over 95% was maintained with leachate addition of 10% and 20% (v/v) while significant decrease of anaerobic treatment efficiency due to inhibition of microbiological activity was observed at leachate concentration above 30% (v/v).