3.1. Characteristics of food waste and inoculum
The pH, organic content (TS & VS), COD, and macromolecular components (protein & carbohydrate) of food waste and inoculums were measured, and the results were presented in Tables 2 and 3.
Table 2
The characteristics of food waste
Characteristics | This study | [40] | [2] | [35] |
pH | 6.2 | - | 4.91 | 6.7 |
Total solid, TS (g/L) | 85.67 | 158 | 66 | 295 |
Volatile solid, VS (g/L) | 77.78 | 151 | 63 | 280 |
VS/TS | 0.91 | 0.96 | 0.96 | 0.95 |
Chemical oxygen demand, COD (mg/L) | 68750.00 | 157000.00 | 110000.00 | 394000.00 |
Total Protein (mg/L) | 4403.33 | - | - | - |
Soluble Protein (mg/L) | 3890.33 | - | - | - |
Total Carbohydrate (mg/L) | 36074.00 | 99000.00 | - | - |
Soluble Carbohydrate (mg/L) | 21034.67 | 54000.00 | - | - |
Table 3
The characteristics of inoculum
Characteristics | This study | [40] | [42] | [41] |
pH | 7.80 | 7.50 | 7.9 | 7.8 |
Total solid, TS (g/L) | 36.78 | 26.70 | 39.0 | 79.06 |
Volatile solid, VS (g/L) | 20.78 | 18.90 | 26.60 | 62.74 |
VS/TS | 0.56 | 0.71 | 0.68 | 0.79 |
Chemical oxygen demand, COD (mg/L) | 30000 | - | 29200.00 | - |
Total Protein (mg/L) | 3789.67 | - | - | |
Soluble Protein (mg/L) | 1910.33 | - | - | |
Total Carbohydrate (mg/L) | 4680.67 | - | - | |
Soluble Carbohydrate (mg/L) | 1600.33 | - | - | |
Food waste used in this investigation appears to have an acidic pH. Li et al. [33] investigated anaerobic digestion of food waste, and the pH of the waste employed in his study was also acidic. Zhang et al., [34], also reported a pH of 6.5 for the food waste utilised in the experiment, which is acidic. The food waste employed as a substrate for anaerobic digestion was acidic, as shown in Table 2.
TS represents the organic content of the substrate while the VS represents the biodegradable volatile solids and refractory volatile solids [35]. VS/TS ratio was determined to express the organic contents of the matter [35]. The total solid and volatile solid content of food waste reported in this study is slightly higher than that of Pramanik et al., [2], but lower than that of Rajagopal et al., [36]. Table 2 shows that the VS/TS ratio of all food waste utilised is greater than 90%. According to Qiao et al., [35], the VS/TS ratio of food waste utilised in anaerobic digestion was typically 80 percent or higher.
COD was conducted on the substrate to measure the amount of oxygen needed to oxidised matter [32]. The COD concentration observed in this study was 68 750 mg/L. Wijayanti et al., [37] stated that feedstock which the COD concentration above 20 000 mg/L is considered as a high degradable feedstock. Food waste had a high COD concentration in this study, indicating that it is a highly degradable feedstock. Table 2 shows the COD content of food waste in several experiments, indicating that food waste is a highly degradable feedstock. High degradable feedstock is suitable to be used in anaerobic digestion as it is easier for the matter to be converted into biogas [38].
The testing for carbohydrate and protein were conducted in total and soluble form. The total and soluble form of food waste represents the particulate fraction and soluble fraction of food waste [39]. The carbohydrate content of food waste and inoculum reported in this study is higher than the protein level. Yeshanew et al., [39], reported a similar pattern in the anaerobic digestion of food waste. In comparison to this study, Yeshanew et al., [39] found a larger level of carbohydrate and protein content.
The pH of the inoculums employed in this work is alkali, as shown in Table 3, and the inoculum was taken from an anaerobic digester processing palm oil mill effluent (POME). POME is a highly polluted wastewater if it is directly discharge without any treatment to it [40]. Kumar et al., [41], stated that the pH of inoculums utilised for anaerobic digestion of food waste was 7.5, and that the inoculum was acquired from a wastewater treatment plant. Parra-Orobio et al., [42] and Forster-Carneiro et al., [43] investigated anaerobic digestion of food waste using inoculum from a wastewater treatment plant. Inoculum pH was 7.8 and 7.9, according to Parra-Orobio et al., [42] and Forster-Carneiro et al., [43]. The pH level that is measured may vary. The inoculum employed in the anaerobic digestion of food waste ranges from 7.5 to 7.9, according to the data in Table 3. In this investigation, the pH of the inoculum was 7.8.
The total solid and volatile solid content of the inoculums employed in this investigation were higher than those reported by Kumar et al., [41]. The TS and VS results in this study is quite similar with the results observed by Forster-Carneiro et al., [43]. Wang et al., [7] stated that a biomass with VS/TS above 50% is considered as a biomass with high organic content. This study obtain a VS/TS ratio of 56% which indicates the inoculum used was slightly rich in organic content. Based on the results tabulated in Table 3, all the inoculums has high organic content and the inoculums are used in the anaerobic digestion of food waste.
The inoculum used in this study resulted in COD concentration of 68 750 mg/L. The result obtained in this study is slightly lower than the COD concentration reported in Chan et al., [30]. Chan et al., [30] studied anaerobic digestion by using POME as inoculum and the COD concentration reported was 70 000 mg/L. Chan et al., [30] also specified a COD concentration range in inoculum that could be used in the anaerobic digestion process. The COD content of inoculum must be between 65 900 and 85 300 mg/L [30]. The COD concentration of the inoculum employed in this investigation remained within the acceptable limit for anaerobic digestion. According to Table 3, the measured COD concentration is lower than the reported COD concentration in other literatures. Despite that, all the reported COD concentration were higher than 20 000 mg/L indicating that the inoculum usually used in the anaerobic digestion process is also highly degradable.
3.2.Characteristics of influent
Table 4 shows the characteristics of the influent for this study. The mixture of food waste and inoculum fed into the batch pilot-scale anaerobic digester is referred to as influent in this study. The pH of the influent was measured at 7.2, which is within the anaerobic digestion process's permitted range. According to Saragih et al., [23], the ideal pH range for anaerobic digestion is between 7 and 8.5. In addition the pH of mixture ranging from 6.8–7.4 is also acceptable [13, 44]. The COD of influent in this study is slightly higher by 2.04 g/L than what reported by Alizadeh et al., [45]. The VFA of influent in this study was below 500 mg/L. Chan et al., [30] also observed a VFA value for influent to be below 500 mg/L and the VFA in this study is considered to be low and less inhibition may occur.
Table 4
The characteristics of influent
Characteristics | Influent |
pH | 7.2 |
Total solid, TS (g/L) | 48.44 |
Volatile solid, VS (g/L) | 32.33 |
VS/TS | 0.67 |
Chemical Oxygen demand, COD (mg/L) | 44266.67 |
Alkalinity (mg CaCO3/L) | 150.00 |
Volatile fatty acid, VFA (mg/L) | 118.67 |
3.3.Process stability
The pH, VFA concentration, and VFA/TA ratio were all evaluated to determine the stability of the anaerobic reactor [46]. On a daily basis, pH, VFA, and the VFA/TA ratio were measured and monitored [3, 32]. Daily effluent was collected and mixed before monitoring tests were performed [2]. The monitoring process were conducted to overcome early instability [33].
Throughout the anaerobic digestion process, the monitoring parameters will be evaluated to detect any inhibitions and to maintain a stable anaerobic digestion process. For 26 days, the pilot-scale anaerobic digestion of food waste was run constantly. Figure 2 depicted a graph of pH and VFA concentration versus day.
As depicted from Fig. 2 throughout the anaerobic digestion process the pH regularly changes. The lowest and highest pH recorded during the anaerobic digestion process were 6.71 and 7.45. Despite the changes in pH, the pH of the effluent was still within the optimum range which is from 6.30 to 7.80 [47]. The pH began to drop from day 1 to day 2, and the VFA concentration remain constant for 6 days. This is due to the high amount of VFA produce in the anaerobic digester in which during the early digestion process, many organics in food waste is rapidly converted into VFA causing the pH to drop [6, 48]. Then the pH start to increase from day 3 to day 7 while the VFA concentration drop significantly at day 7. According to Pramanik et al., [2,] the rising pH was caused by a low VFA concentration and a high ammonia concentration. From day 8 to 9, the pH drops, then rises until the anaerobic digestion process is completed. When the protein in the food waste is degraded, it also resulting in high ammonia concentration, high amount of hydrogen sulphide, as well as producing long chain of fatty acid, causing the pH to increase [6, 49]. In addition the increase pH may resulted because of the digester foaming when the food waste is digest [50].The VFA concentration began to decrease from day 8 to day 26. Although VFA concentrations began to rise on days 15 and 16, they remain low when compared to VFA concentrations during the first eight days of the anaerobic digestion process. The low VFA concentrations indicate a high biomethane conversion efficiency [51]. In this investigation, the final VFA concentration was low, and the final pH was within the ideal range. A similar trend has been observed by Li et al., [51], when the final VFA content is low and the final pH is suitable at the end of the anaerobic digestion process.
Table 5 tabulates the measured pH and VFA/TA ratio for 26 days. VFA/TA is also used as a tool in analysing the stability of the anaerobic digestion process along with pH, and VFA [33]. The highest VFA/TA ratio obtain is at day 2 and day 5 where early digestion stage takes place in which VFA is produced when the food waste is degraded. The highest VFA/TA ratio obtained in this study was 0.35. The anaerobic digestion process is deemed stable when the VFA/TA ratio is less than 0.35, according to Chan et al., [30], Li et al., [33], and Li et al., [52]. When the VFA/TA ratio is less than 0.4, according to Raposo et al., [27], the anaerobic digestion process works successfully without the risk of acidification. The highest VFA/TA ratio reported during this investigation was 0.35, which was still within the optimal range, indicating that the anaerobic digestion process remained steady for the whole 26-day period. The digester remain stable throughout the anaerobic digestion process aid in active microbial activity thus enhancing methane production.
Table 5
The pH and VFA/TA ratio throughout the anaerobic digestion process
Days | VFA/TA | Days | VFA/TA | Days | VFA/TA |
1 | 0.33 | 11 | 0.08 | 21 | 0.08 |
2 | 0.35 | 12 | 0.08 | 22 | 0.08 |
3 | 0.34 | 13 | 0.08 | 23 | 0.08 |
4 | 0.34 | 14 | 0.08 | 24 | 0.08 |
5 | 0.35 | 15 | 0.15 | 25 | 0.08 |
6 | 0.30 | 16 | 0.14 | 26 | 0.08 |
7 | 0.24 | 17 | 0.08 | | |
8 | 0.16 | 18 | 0.08 | | |
9 | 0.17 | 19 | 0.08 | | |
10 | 0.14 | 20 | 0.12 | | |
3.4.Pilot-scale anaerobic digester performance
Next removal efficiency was studied to visualize the rate of the organic degradation throughout the anaerobic digestion process. The removal efficiency was calculated using the parameters TS, VS, and COD, where TS, VS, and COD indicate the solid content of the food waste [34]. Figure 3 depicts the removal efficiency data collected every day for 26 days.
The removal efficiency was estimated to determine the rate of organic decomposition and to assess the anaerobic digestion process' effectiveness [2]. The food waste inside the digester is decomposed and biogas is created during the anaerobic digestion process, generating changes in TS, VS, and COD concentrations [2]. The maximum removal efficiency for TS, VS, and COD in this investigation was 85.32, 94.15, and 93.52 %, respectively. Lin et al. [53] used a single-stage anaerobic digestion of food waste to achieve the greatest COD removal rate of 74 %. Kumar et al., [41] also claimed to have achieved the greatest COD elimination rate of 74 %. This study obtained greater COD removal than Lin et al., [53] and Kumar et al., [41]. Pramanik et al. [2] investigated single-stage anaerobic digestion of food waste and measured the average TS, VS, and COD removal efficiency. Pramanik et al. [2] observed removal rates of 72.20, 78.90, and 80.0 % for TS, VS, and COD, respectively. The average removal of TS, VS, and COD in this investigation was 78.39, 85.95, and 86.23 %, respectively. In this study, the average removal of TS, VS, and COD was slightly higher than the results obtained by Pramanik et al., [2].
The results of this study's TS, VS, and COD removal efficiency showed that there were active microorganisms present in the inoculum, in which the food waste was digested and anaerobic digestion was carried out [54]. In this study, COD resulted has the highest average organic degradation followed by VS, and TS. In addition, similar with what observed by Pramanik et al., [2] the COD removal has the highest average removal followed by VS removal and lastly TS removal.
3.5.Methane Accumulation
Figure 4 depicts the accumulation of methane during the anaerobic digestion of food waste. In this investigation, the methane accumulation was monitored for 26 days. The experiment ceased when the methane production was consistently low and reach plateau. For this study, the methane accumulation started to reach the plateau at day 19. The methane accumulation for this study was 463250 mL or 463.25 L respectively. Methane was increasingly produced proportional with the organic degradation [33]. High organic degradation proven an active microbial activity thus more organic is converted into methane each days [33]. Stable anaerobic digestion system with low inhibitions contributes to high methane production [11].
3.6.Methane Yield
The methane yield is measured to determine the effectiveness of anaerobic digestion of food waste [2]. The methane yield for anaerobic digestion of food waste is tabulated in Table 6. Methane gas production begins on day 1 according to the statistics shown. The methane gases start to produce on day 1 similar with what reported by Huang et al., [55] and Li et al., [51]. For this study, the methane production increase from day 1 until day 10 and then the methane production slowly decreases until day 26. The ultimate methane yield for this study was 5103.6 mL CH4/gVS respectively.
Table 6
The methane yield (mLCH4/gVS) for anaerobic digestion of food waste
Days | Methane yield (mLCH4/gVS) | Days | Methane yield (mLCH4/gVS) | Days | Methane yield (mLCH4/gVS) |
0 | 0 | 10 | 3503.4 | 20 | 4830.9 |
1 | 293 | 11 | 3864.7 | 21 | 4881.0 |
2 | 837.8 | 12 | 4122.5 | 22 | 4925.1 |
3 | 1314.9 | 13 | 4247.5 | 23 | 4968.1 |
4 | 1649.8 | 14 | 4363.2 | 24 | 5009.9 |
5 | 1893.2 | 15 | 4462.4 | 25 | 5060.6 |
6 | 2203.9 | 16 | 4571.4 | 26 | 5103.6 |
7 | 2479.9 | 17 | 4659.6 | | |
8 | 2823.6 | 18 | 4727.3 | | |
9 | 3103.4 | 19 | 4786.8 | | |
This study obtained high removal efficiency of TS, VS, and COD (84.64, 92.79, and 93.52%), which were above 50% indicating high organic degradation [33]. The methane yield of food waste may be influenced by the rate of removal efficiency [33]. The removal effectiveness of TS, VS, and COD increased the methane production dramatically [33]. The carbohydrate content of the food waste used in this investigation is higher than the protein content. When food waste rich in carbohydrate was degraded, it tend to produce high methane yield [38].
According to the VFA/TA ratio, this study remains stable throughout the digestion process and providing a suitable working environment for the microorganisms to produce methane [12]. In this study, the anaerobic digester was semi-continuously mixed for 30 minutes at 70 rpm [2]. In a semi-continuously mixed digester, high methane yield can be obtained [57, 58]. Semi-continuously mixing during the anaerobic digestion process can provide effective microbial activity that can enhanced the methane production [29, 58].
3.7.Kinetic Analysis
The performance of anaerobic digestion of food waste was described using kinetic analysis [2]. Modified Gompertz (GM) modelling and a Logistic Function Model were used in this investigation.
Table 7 lists the expected parameters for anaerobic digestion of food waste, such as ultimate methane yield, methane production rate, and lag phase. The graph of methane yield between real data and estimated data from Modified Gompertz modelling and Logistic function modelling for this investigation is shown in Figs. 5 and 6. The figures depicted can be explained and divided into three section [2]. The first section elaborates the relationship of lag phase to establish the biogas production [3]. The second section was the exponential phase where the biogas production will be peak [3]. Lastly the third section is called death phase or steady phase where the biogas production became lesser and the graph became plateau [2].
Table 7
The kinetic analysis of anaerobic digestion of food waste
Model | Kinetic parameters | Laboratory (Actual) | Estimated |
Modified Gompertz | Ultimate methane yield (mlCH4/gVS) | 5103.56 | 4858.10 |
Methane production rate, (mlCH4/gVS/day) | 293.00 | 453.09 |
Lag phase (day) | 0.00 | 0.87 |
R2 | - | 0.85 |
Logistic function | Ultimate methane yield (mlCH4/gVS) | 5103.56 | 4632.25 |
Methane production rate, (mlCH4/gVS/day) | 293.00 | 526.54 |
Lag phase (day) | 0.00 | 1.84 |
R2 | - | 0.79 |
For this study in the first section according to the Modified Gompertz modelling, this setup required approximately 1 day of lag phase for methane production. While Logistic function modelling required about 2 days of lag phase for methane production. The second section display the sharp increase of methane production from day 2 to day 17 for modified Gompertz modelling. Meanwhile, the methane production significantly increase from day 2 to day 15 for logistic function model. The significant increase during this period were due to the rapid development of microbial community and effective degradation process [2]. The steady phase (Sect. 3) was from day 18 to day 26 and from day 16 to day 26 for modified Gompertz model and Logistic function model correspondingly. In the steady phase, the methane production was gradually slow and slowly stopped due to the reduction of active microbial community and low degradation process [56]. The estimated data from Modified Gompertz modelling and Logistic function model were lower than the actual experimental data. These results were similar with what reported in Pramanik et al., [2].
The experimental data and kinetic parameters of the Modified Gompertz model and Logistic function model [2] were used to determine the methane production rate. A high R2 value (above 0.5) suggests that the model used fits the experimental data well [2]. The modified Gompertz model and the Logistic function model used in this study had excellent R2 values of 0.85 and 0.79, respectively, and were well fitted to the experimental data. The modified Gompertz model had a higher R2 value than the logistic function model in this investigation. This demonstrated that the modified Gompertz model provided a more exact estimate and could explain more than 85% of the variation in the results [2]. According to earlier studies [16, 59], the modified Gompertz model fit the data better than the logistic function model.