Physicochemical properties of the samples are shown in Table 1. The ash, TS and VS were determined to be 14.22%, 97.00% and 82.79% for Hura crepitans leaves and 5.42%, 30.14% and 24.72% for the inoculum. The C/N ratio Hura crepitans leaves was 32.2 while its lignin, cellulose and hemicellulose contents were measured to be 28.26%, 32.70% and 39.05% respectively. As earlier mentioned, Hura crepitans leaves were thoroughly mixed with the inoculum and water added to produce the three F/I ratios used in the DAD.
Sample DAD6 had the highest lignin content while DAD2 had the least. Lignin usually forms a complex network with the carbohydrates thus making the lignocellulosic biomass recalcitrant to microbial degradation, which in turn could lower biogas yield [9].
Both samples DAD2 and DAD4 had higher contents of hemicellulose and cellulose than DAD6 which are the major components from which biogas is produced [20]. The highest ash content was found in sample DAD2 followed by DAD6. The cellulose content for DAD2 was close to 51.30% reported for organosolv-pretreated pinewood at 150 oC for 30 minutes by Mirmohamadsadeghi [21]. The hemicellulose content on the other hand was higher than 19.1% reported for corn cob by Liew et al [22].
The C/N ratios of DAD4 and DAD6 were well outside the range recommended for AD, that is, between 20:1 and 30:1 [23] while DAD2 was slightly above (30.95%) the range. The C/N ratios of DAD4 and DAD6 were however, below 97.6% reported for sunflower stalk by Kurt et al. [24]. Despite the high C/N ratios (with the exception of DAD2) there was still reasonably high biogas production. This observation has been reported in literature [25, 26] which could be attributed to the fact that C/N is not the only parameter responsible for effective running of AD. The overview done by Deublein and Steinhauser (2011) have pointed out no direct correlation between C/N ratio and biogas yield. [27] Their presumable conclusion was that C/N ratio might not in any way affect biomethane (biogas) yield. They based their argumentation on a three-year study on buffalo manure collected under varying conditions with C/N ratios ranging from 9 to 50. Their findings showed that there was significant biomethane production outside the range often reported in literature as optimal for AD. The argument can be laid to rest when we consider the fact that there are other factors that contribute to biogas production beside C/N ratio. There is need for further research in this area because of the contradictory results that have been reported in literature in order to have a better understanding of the role of C/N ratio on AD. However, substrates high in C/N ratio, such as samples DAD4 and DAD6 could serve as good co-digestion feedstocks for substrates that are low in carbon content [27].
Table 1
Physicochemical properties of the samples used in dry anaerobic digestion
Parameters | Hura crepitans leaves | Inoculum | DAD2 | DAD4 | DAD6 |
Ash content (%) | 14.22±0.15 | 5.42±0.31 | 15.32±0.17 | 13.54±0.76 | 14.29±0.54 |
TS content (%) | 97.00±0.05 | 30.14±0.05 | ND | ND | ND |
VS content (%) | 82.79±0.15 | 24.72±0.3 | ND | ND | ND |
Carbon content (%) | 47.66±0.08 | ND | 47.05+0.05 | 47.73+0.42 | 47.62+0.30 |
Nitrogen content (%) | 1.48+0.28 | ND | 1.52+0.83 | 0.63+0.04 | 0.57+0.07 |
C/N ratio | 32.20 | ND | 30.95 | 75.76 | 83.54 |
Lignin (%) | 28.26±0.74 | ND | 26.77+0.53 | 36.48+0.04 | 44.50+0.54 |
Cellulose (%) | 32.70+0.34 | ND | 50.33+0.04 | 45.76+o.29 | 38.58+0.5 |
Hemicellulose (%) | 39.05±0.39 | ND | 22.91+0.55 | 17.75+0.25 | 16.94+1.03 |
ND − not determined |
The pH values of the initial and final samples are shown in figure 2. The values ranged from 5.4-8. The initial pH of F/I 2 was 8.00 while the final pH was 5.60. For F/I 4 the initial pH was 7.60 while the final pH was 5.40. Digester F/I6 also had initial pH of 7.60 but a final pH of 6.00. Overall, there was a decline in pH values in all the digesters at the end of the anaerobic digestion, a fact that has been attributed to accumulation of volatile fatty acids [28].
Daily And Cumulative Biogas Production
Figure 3 shows daily and cumulative biogas production from the three weighted samples. There was no lag phase in all the digesters as production commenced within the first 24 hours which was indicative a good start up/hydrolysis stage in the DAD process. It has been observed that operating DAD at a higher F/I ratio could slow down the initial stage as shown elsewhere [10] but the results from this study revealed that the F/I ratios adopted were appropriate for rapid beginning phase of the DAD. Digester DAD6 peaked earlier (day 7) than other digesters with peak production of 280 ml. Digester DAD2 peaked on day 8 with 410 ml while digester DAD4 achieved its highest peak value on day 18 (690 ml) which was longer than other digesters. Biogas continued till the last day of the experiments which shows that the process efficiency was healthy which could be as a result of no or low inhibitory substances in the system.
In this study, the sample DAD6 with highest lignin content produced the least amount of biogas. Although no clear relationship between the amount of biogas produced and lignin contents of the substrates in this study, yet the least biogas produced was from the sample with the highest lignin content.
Results of cumulative biogas production revealed that digester DAD4 was 2.4 fold higher than that of DAD6 which produced the least volume (940 mL) of biogas. Digester DAD2 produced a volume of 2005 ml while digester DAD4 produced a highest volume of gas with a total of 2210 ml. This shows that biogas production is best carried out at this F/I ratio.
Kinetic Study
There was a direct correlation between biodegradability constant and biogas yield. DAD4 with the highest biogas yield had the highest constant biodegradability value of (Ko) of 1.110 d−1 followed by DAD2 0.569 d−1 with sample DAD6 having the least value of 0.347 d−1. The results revealed that digester DAD6 had highest utilization of the substrate, which translated to increased biogas production compared to other digesters. The Ko for DAD2 was higher than 0.9185 reported for 3% Organosolv pretreated rice husk by Olugbemide et al. [17].
Table 2
Chemical kinetics parameters regressed for various proposed models/data.
Data
|
DAD2
|
DAD4
|
DAD6
|
Exponential
|
|
|
|
Pm/L
|
2.05
|
2.21
|
1.01
|
k /day−1
|
0.146
|
0.064
|
0.094
|
Logistic
|
|
|
|
Pm /L
|
1.92
|
2.58
|
0.90
|
Rm /(L day−1)
|
0.272
|
0.124
|
0.066
|
λ /day
|
1.80
|
5.72
|
0.24
|
Gompertz
|
|
|
|
Pm /L
|
1.94
|
3.42
|
0.92
|
Rm /(L day−1)
|
0.273
|
0.102
|
0.068
|
λ /day
|
1.58
|
3.45
|
0.12
|
Results show that exponential kinetic model provides the poorest measurement agreement, not being able to describe the lag in production, which is clearly seen, in particular with the data 2, where maximal production rate is achieved only after day 17. Logistic or Gompertz models provide approximately the same prediction/ measurement agreement, whereas the data 1 or 3 can be described excellently, while for 2, we not only see a lag in production, but also a bimodal digestion process. Correspondingly, this hints towards two predominant fractions, which are being digested at a much dissimilar rate/lag of cascade digestion process, which should have been described by a reaction in series kinetic model, regressing also intermediate.