Improving the biogas production in two-phase anaerobic digester of food waste using sugarcane bagasse-derived biochar

The OFMSW was compiled from the rubbish of different houses in a residential zone in Cairo, Egypt. The components of OFMSW were the residuals of meat, beans, bread, vegetables, and rice. After removing leftovers such as bones, OFMSW was softened to small particles by an electrical blender. Then, the resulting slurry was sieved using a stainless-steel sieve with 2.0-mm openings. Subsequently, the slurry was saved in 1000-L plastic drums in a refrigerator at − 4 °C to keep the feedstock from degradation at room temperature and prevent any changes in the feedstock’s composition during the experiments.

The anaerobic sludge (AS) was sourced from an anaerobic digestion facility in El-Gabal El-Asfar Wastewater Treatment Plant in Al Qalyubia, Egypt. The collected anaerobic consortia were left for 24 h to decant the supernatant, and the concentrated sludge was collected. The concentrated sludge was also sieved to remove the large particles and kept under anaerobic conditions for 1 month before the experiments. Table 1 shows the characteristics of the substrate and inoculum.

All parameters were measured twice, and the table provides the average values. TCOD, SCOD, TSS, NH₃-N, and VS refer to the total chemical oxygen demand, soluble chemical oxygen demand, total suspended solids, ammonia–nitrogen, and volatile solids.

The sugarcane bagasse wastes were compiled from sugarcane juice shops. Then, they were washed with distilled water and air dried for 24 h. Subsequently, they were cut into small pieces, and the small pieces were placed in a muffle furnace purged with nitrogen (100 mL/min) at 500 °C for 2 h. The obtained particles were washed with HCl and water and dried in an oven for 24 h at 105 °C.

The two-stage batch AD process was performed in glass flasks with a total volume of 500 mL and a working volume of 300 mL. The feedstock and inoculum volumes were 150 mL and 150 mL with a volumetric ratio of 1:1. For the first stage, the sludge was initially preheated at 70 °C for 30 min to inactivate methanogens or other H₂-consuming microorganisms, and the pH was adjusted to 5.5 using 1.0 M of NaOH or HCl in the acidogenic reactor. All the reactors were purged with nitrogen for 10 min to confirm the anaerobic conditions, and then they were tightly sealed. Further, all the reactors were kept at 37 °C during the experiments and agitated at 200 rpm for 10 min every 12 h. The produced hydrogen from the bioreactors was measured via the water displacement method until the biogas production greatly decreased or stopped. Then, 150 mL of unheated sludge was added to the digested mixture to initiate the methanogenic stage. The pH was then adjusted to 7, and the produced methane was measured using the water displacement method.

A control experiment was performed using the OFMSW and the inoculum without adding the synthesized biochar. To evaluate the effect of biochar, three different doses (5 g/L, 10 g/L, and 15 g/L) were added in both stages to identify the optimal dose based on the maximum hydrogen and methane produced. All the experiments were duplicated to affirm the accuracy of the obtained results. Volumes and composition were monitored daily during the experiments. Volatile fatty acids (VFAs), pH, and alkalinity were measured every 5 days. Additionally, TSS, VS, TCOD, SCOD, alkalinity, and ammonia–nitrogen (NH₃-N) were measured after the 30 days of batch operation.

The morphology and chemical constituents of the biochar were studied using scanning electron microscopy (SEM) (Joel JEM 2100, Japan) and elemental analyzer, respectively. Also, the chemical structure and functional groups of the biochar were investigated through X-ray diffraction (XRD 6000, Shimadzu, Japan) and Fourier transform infrared spectroscopy (Shimadzu, FTIR-8400S). Further, the surface area and pore size of the biochar were explored using Belsorp-max automated apparatus. TCOD, SCOD, NH₃-N, alkalinity, TSS, and VS were measured according to the procedures mentioned in the Standard Methods for the Examination of Water and Wastewater [22]. The composition of the biogas collected from the reactor was determined using a biogas 6000 analyzer.

The main interaction, correlation, and effect plots were developed via Minitab and R.

VFAs’ concentrations in the liquid samples withdrawn every 5 days were measured by high-performance liquid chromatography (HPLC 1260 series). The detection was performed via an AQ-C18 HP column (4.6 mm × 150 mm, particle size = 3 µm) with a sample volume (5 µL). 0.005 N sulfuric acid was employed as a mobile phase with a linear gradient, where the flow rate was 0.8 mL/min (0–4.5 min), 1 mL/min (4.5–4.71 min), 1.2 mL/min (4.71–8.8 min), 1.3 mL/min (8.8–23 min), and 0.8 mL/min (23–25 min). The wavelength of the detector and the temperature of the column were 210 nm and 55 °C, respectively.

The SEM image in Fig. 1a depicts the sheet-like structure of the biochar’s particles with high porosity. The surface area of the biochar is 180.72 m²/g. Further, total pore volume and half pore width are 0.1874 cm³/g and 1.53 nm, respectively. The biochar’s high surface area could act as a carrier for microbial growth and biofilm formation, thereby increasing biogas production [23]. Elemental analysis shows that 94.6% of the prepared biochar is carbon, while nitrogen and hydrogen contents are 0.32% and 4.9%, respectively. The broad peaks in the XRD pattern in Fig. 1b between 20°–30° and 40°–50° are imputed to the diffraction planes of (002) and (100), respectively, of the amorphous structure of carbon [24]. The band at 3400 cm⁻¹ in FTIR spectra in Fig. 1c affirmed the existence of the O–H bond, whereas the bands at 1585 cm⁻¹ and 1700 cm⁻¹ are imputed to the stretching vibrations of C = C and C = O, respectively [25]. Further, the band at around 1215 cm⁻¹ is imputed to symmetric stretching of the C-O group [26]. Moreover, the band at 760 cm⁻¹ is allocated to the bending of the C-H group [26]. The oxygen functional groups on the biochar could act as an electron mediator between bacterial species via direct interspecies electron transfer (DIET), which promotes the production of biogas [27].

Bild vergrößern

The TSS, TVS, TCOD, SCOD, NH₃-N, and VFAs were monitored during the operation of bioreactors. The initial features of the mixture are listed in Table 1. TSS values declined to 4350, 3792, 2629, and 8204 mg/L, whereas VS decreased to 3872, 3008, 1937, and 6280 mg/L in the case of control (without biochar), 5 g/L, 10 g/L, and 15 g/L biochar dose, respectively, at the end of the reaction (30 days) (see Fig. 2a). TSS and VS greatly decreased due to the enhanced hydrolysis of complex organics to simpler compounds, which reduced the solids existing in the system. Further, the highest reduction of TSS and TVS was observed in the case of biochar 10 g/L, which could possibly enhance the hydrolysis rates via activating the hydrolases (e.g., lipase, protease, and amylase) [28]. However, the TSS and VSS reduction ratios declined at biochar dose above 10 g/L, as the increase of biochar dose over the optimum could adsorb carbon sources, thereby deaccelerating the microbial activity and hydrolysis [29].

Bild vergrößern

Regarding the TCOD and SCOD, the TCOD decreased to 83.028, 73.196, 62.942, and 93.028 g/L, while SCOD values were 40.184, 35.926, 32.561, and 47.138 g/L in the case of the absence of biochar, 5 g/L, 10 g/L, and 15 g/L biochar dose, respectively as given in Fig. 2b. The observed increase of TCOD and SCOD removal with time was due to the microorganism’s adaptation to the organic load [30]. The highest removal efficiency of TCOD and SCOD was attained in the case of a biochar dose of 10 g/L due to the potential of biochar to reduce the lag phase and act as a carrier of electrons between microbes, thereby improving the biodegradation of organics [23].

The pH value was initially adjusted to 5.5 for the acidogenic phase. Then, the pH was further adjusted to 7 on the 11th day during the start of the methanogenesis phase. The pH values were monitored during the 30 days (see Fig. 2c). The changes in pH were significant in the case of the absence of biochar, whereas the pH values were approximately the same in the presence of biochar due to its buffering capacity [31].

Ammonia and alkalinity are important factors for the stability of anaerobic digesters. Thus, these parameters were monitored during the 30 days (see Fig. (2d, e). The alkalinity decreased in the first phase (hydrogen production, 10 days) due to the consumption of alkalinity for buffering pH because of the generation of high concentrations of VFAs. On the other hand, the alkalinity increased in the second phase (10–30 days) due to the degradation of VFAs and the neutral pH (6.9) in the case of the methane production phase. The NH₃-N concentrations declined to 270, 180, and 220 mg/L in the case of 5 g/L, 10 g/L, and 15 g/L biochar dose, respectively, at the end of the operation (30 days) (see Fig. 2e) due to the consumption of ammonia by microorganisms for growth and reproduction [32]. As suggested in the literature, biochar could also adsorb NH₃-N, which improved the removal efficiency of NH₃-N [33]. On the other hand, the NH₃-N concentration raised to 660.8 mg/L in the absence of biochar (control) due to the decomposition of nitrogen species (e.g., proteins) in the hydrolysis stage [34].

The VFAs such as formic acid, lactic acid, acetic acid, citric acid, succinic acid, propionic acid, and butyric acid were detected during operation (see Fig. 3). The concentrations of the aforementioned acids fluctuated with time. The concentrations of VFAs increased with the addition of the biochar compared to the control in the acidogenic phase, while the degradation of VFAs was observed in the methanogenic phase. The main VFA was lactic acid, with a concentration ranging from 16,036.96 to 24,410.66 mg/L, where butyric acid had the lowest concentration (0–49.4 mg/L). The highest concentrations of VFAs were attained in the first phase (hydrogen production), and faster degradation of VFAs was achieved in the second phase (methane production) in the case of adding 10 g/L of the biochar due to its potential to reduce lag phase [21]. The low concentrations of VFAs confirm their faster degradation to methane production, as the accumulation of VFAs could inhibit the microbial activity due to pH drop [34].

Bild vergrößern

As shown in Fig. 4a, hydrogen production started on day 1 and gradually decreased until the hydrogen production almost stopped on day 10. The methane production was also evident during the remaining 20 days of operation after adding the unheated sludge to the system. Thus, the hydrogen and methane were measured for 10 and 20 days, respectively (see Fig. 4a). Further, the biogas composition in the case of the control experiment and under different biochar doses is provided in Fig. 4b–e. Hydrogen was detected during the first 10 days as the main component of the biogas with a maximum content of 85.6% and maximum yield of 583.2 mL/g VS for the biochar dose of 10 g/L. The cumulative yields of hydrogen were 772.7, 955.9, 1653.8, and 1293.6 mL for the control, 5 g/L, 10 g/L, and 15 g/L of biochar, respectively. The hydrogen volume in the control significantly decreased on the 5th day due to the accumulation of VFAs resulting in a drop in pH to 4.8. However, in the case of biochar, especially a biochar dose of 10 g/L, the hydrogen production was higher due to the stability in pH values. On the 11th day, methane started to prevail. The highest methane yield (114.5 mL/g VS) was detected on the 15th day for 10 g/L biochar dose, and the highest methane content was 81.6%. In the first 10 days, organic acids could be transformed to H₂ and CO₂ via acetogenic bacteria, whereas acetic acid was generated through the homoacetogenic bacteria. After 10 days, the methanogenesis stage could start, where methane was formed through acetic acid via acetoclastic methanogenic archaea, as well as from H₂ and CO₂ through hydrogenotrophic methanogenic archaea. The cumulative methane yields were 960, 1150, 1354.5, and 1204.1 mL/g VS in the case of control, 5 g/L, 10 g/L, and 15 g/L. The improvement of hydrogen and methane production in the case of introducing biochar might be due to the ability of biochar to act as a habitat for microbial colonization [23]. Further, the biochar’s surface could act as a platform for transferring electrons between microbial species which enhances the production of biogas [27]. Additionally, adding biochar could reduce the lag phase and accelerate the hydrolysis by activating the hydrolases such as lipase, protease and amylase [28]. Sugiarto et al. [23] showed that hydrogen yield was improved by 107% after adding the biochar. However, raising the biochar above 10 g/L could decrease the volumes of hydrogen and methane, as overdoses of biochar could adsorb carbon sources which would inhibit the microbial activity, thereby suppressing the biogas production [29, 35]. Further, high doses of the biochar can result in releasing large amounts of alkali-metal and organics which significantly suppressed the microorganisms growth and activity. Shi et al. [28] reported decreased methane yield after adding higher biochar doses above the optimal dose. The low CO₂ content might be due to the potential of the biochar to capture CO₂ due to the high aromaticity and hydrophobicity of biochar-based materials as reported in the literature [36, 37].

Bild vergrößern

The statistical technique principal component analysis (PCA) was employed to investigate the interrelations of various aspects in AD process supported and not supported with biochar. As a useful multivariate analytical approach, PCA analyzed the correlations between AD variables while reducing the original set of variables to specific number of principal components (PCs) [38]. Our set of observations and variables, that are possibly correlated, was translated to PCs to attain a set of values that are linearly independent and assist in the decision-making.

To identify the influence of different AD schemes on the variations of VFAs production and the relative performance, PCA was studied. The first investigation is depicted in the PCA plot (Fig. 5). Two main groups were distinguished as illustrated in score plot (Fig. 5a). The first group in the top right includes four samples for control and biochar 5, 10, and 15 g/L doses at day 15. The second group in the top left includes two samples and biochar 5 and 15 g/L doses at day 30. The other distinct samples that do not belong to any clusters are biochar 10 g/L at day 30, control at day 30, and condition. The PCA results are consistent with the previous section that highlighted the major variations in VFAs concertation of biochar 10 g/L sample at day 30 in comparison to control and initial concertation. The PCA confirmed the close similarity between the two biochar 5 and 15 g/L doses at day 30 (second group top left) compared to 10 g/L. The four samples are control and 5, 10, and 15 g/L biochar doses at day 15 (first group top right); PCA also indicated that the major differences in VFAs concertation mainly occurred after 30 days.

Bild vergrößern

The PCA biplot (Fig. 5b) describes the similarity in the changes of the concertation between the VFAs mixture in particular four types: lactic, acetic, formic, and succinic acids (i.e., observed in the close vectors of the four acid classes). Compared to butyric acid, both propionic and citric acid vectors were grouped together in same quadrant. As given earlier in Sect. 3.1, the lactic acid (concentration range = 16,036.96 to 24,410.66 mg/L) was the main VFA, where butyric acid had the lowest concentration (0–49.4 mg/L) in which highest concentrations of VFAs were attained in the first phase due to hydrogen production (< 11 days). PCA results match the previous observations in which butyric acid has the highest concertation variations (increase of 100% at day 30) between all VFAs mixture (butyric acid concentration = 0–49.4 mg/L), while the variations in the mixture of lactic, acetic, formic, and succinic acids were second (reduction at day 30 ranged from 20 to 37%) and then followed by differences in both propionic and citric acids (reduction range at day 30 = 7–14%).

We established another PCA check for the same AD treatment conditions (3 biochar doses, initial and control) (Fig. 6). In comparison to the treated biochar samples, the vector of the initial condition sample was individually distinguished from other vectors. Five vectors were distinguished in the top quadrant which includes two samples for control at day 15 and day 30 and three sample for biochar doses 5, 10, and 15 g/L doses at day 15. With the continuous degradation of VFAs and methane production at day 30, three vectors were grouped in the bottom quadrant. Although the three vectors for biochar doses 5, 10, and 15 g/L doses at day 30 are observed in same quadrant, the two samples biochar 5 and 15 g/L doses are close, and the biochar 10 g/L at day 30 was distinct vector. This is similar to the above noted PCA in Fig. 5 in which biochar 10 g/L at day 30 does not belong to any clusters. The relative differences verified by the PCA withstand the relative efficiency and the relative performance of AD conditions with biochar doses in enhancing biogas production during the two AD stages. In summary, PCA proved the close similarity of performance in terms of the biogas production during the two AD stages for the two biochar doses 5 and 15 g/L at day 30 (in the same quadrant and same group top left, Fig. 5a) compared to biochar 10 g/L.

Bild vergrößern