| dc.description.abstract | Predicting biogas yields from the composition of biomass feedstocks has been difficult primarily due to the variations found in their chemical composition and structure of lignocellulosics. This study aims at understanding, assessing and predicting biogas yields from the chemical composition cutting across different types of biomass feedstocks: fruit peels and rinds, cereal straws, crop stalks, husks, leaves and conifers. A wide range of substrates were chosen to account for the variation in composition by choosing these feedstocks with large variations in the lignin (range, type and steric arrangement). This research answers a few fundamental questions pertaining to predicting biogas yields, the role played by extractives and lignin and also the arrangement of the constituents.
A classical Biochemical Methane Potential (BMP) approach was taken up to understand the fermentation behavior of various feedstocks and understand the issues/ problems in a systematic manner. The first part of this study deals with predicting the pattern of biogas production. The results show that the biogas yield is a strong function of extractive , i.e., pectins, soluble carbohydrates that leach out into the water during the hot water extraction, content in the biomass species. The feedstocks with high concentration of extractives produced higher amount of biogas and at higher rates. As the lignin concentration rises the extent and rates of gas production falls down. Based on the preliminary results and prior understanding, a new methodology was developed where the gas production can be ascribed to sequential/ semi-sequential degradation of biomass components. The sequential decomposition leading to biogas production can be explained by a two-component fit, wherein the first stage of gas production is ascribed predominantly to the breakdown of the easily accessible extractives and hemicellulose (unbound). In the second stage the gas production is slower and can be ascribed to the conversion of difficult to degrade hemicellulose (bound) and cellulose. The point where the first stage and second stage join has been termed as the “inflection point”. The inflection point was found to be ≈20 days in case of easy to digest feedstocks, that are usually rich in extractives. As the lignin concentration increases the time taken to achieve inflection increases, in this case ≥30 days. The concept of two-component fit and inflection point also shows that ≈4/5th of the total gas production is evolved till it reaches the inflection point.
In a first, a classification-based approach was taken up to arrive at regression equations correlating the biogas yields and composition of the biomass. The 26 feedstocks studied were grouped into three classes based on: a) the degradability in nature and cross verified with b) statistical K-means clustering. The classes obtained were
1. Extractive rich peels: that degrade faster.
2. Lignin rich leaves: that have slow degradability rates.
3. Holo-cellulose rich straws, stalks, husks and grasses: degradation rates at levels in between the above two classes mentioned.
Based on this classification two multiple regression equations were developed. The equations showed strong correlation between the biogas yields and chemical composition with a correlation coefficient of ≥0.95. The equation shows the significance of extractives and hemicellulose in the gas production. In contrast, lignin and ash hinder the gas production. A new indicator L/E (L= lignin and E= extractives) has shown that these two components play a significant role in understanding the AD process. The correlation between gas yields and L/E follows power law with a R2 of 0.70 and needs more refinement to incorporate additional recalcitrants such as waxes, silica, etc.
Lignin has been extensively used as a marker as it has been considered not to degrade under the typical anaerobic environment of an AD system. The correlation between gas yields and lignin has always been considered to follow a negative linear relationship. This study showed that the gas yields follow a power law with lignin as an independent variable. The relationship shows that at some point (in this case 15-20%) the extent of access limitation becomes significant, finally leading to decreased rate and extent of gas production. In order to study the effect of extractive concentration on the AD process, substrate with high extractive was chosen and the inoculum concentration was varied. At low levels of inoculum, the gas production was inhibited due to the inability to take up and rapidly convert the VFA flux (pH was found to be low). On the other hand, when the lignin concentration is high (30%) the degradation is low and methanogens are starved, hence low methane production takes place. Such feedstocks do not show sensitivity to S/I ratios. In this study it was termed as the access limited fermentation. It shows that when lignin concentration is higher the propensity to eclipse underlying hemicellulose and cellulose becomes higher thereby reducing access to enzyme hydrolysis and finally biogas production.
Here lignin is considered as an ‘obstructor’ to access and degradation, an effort has been made to understand the biogas production after simple pretreatment like autoclaving with and without alkali. The classification developed earlier was used to understand the effect of alkali treatment which selectively delignifies the biomass. It was found that a simple autoclaving is sufficient for high gas yields compared to the alkali pretreatment, although alkali pretreatment gives high gas yields in some cases, the loss of volatile solids is about 40-50% during the alkali pre-treatment. Therefore, the delignification step to make cellulosics more accessible, needs to sacrifice a large fraction of extractables that would have contributed to biogas production. | en_US |
