Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality
As one of the world’s largest palm oil producers, Malaysia encountered a major disposal problem as vast amount of oil palm biomass wastes are produced. To overcome this problem, these biomass wastes can be liquefied into biofuel with fast pyrolysis technology. However, further upgradation of fast py...
| Main Author: | |
|---|---|
| Format: | Thesis (University of Nottingham only) |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/72215/ |
| _version_ | 1848800724961460224 |
|---|---|
| author | Chong, Jia Wen |
| author_facet | Chong, Jia Wen |
| author_sort | Chong, Jia Wen |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | As one of the world’s largest palm oil producers, Malaysia encountered a major disposal problem as vast amount of oil palm biomass wastes are produced. To overcome this problem, these biomass wastes can be liquefied into biofuel with fast pyrolysis technology. However, further upgradation of fast pyrolysis bio-oil via direct solvent addition was required to overcome it’s undesirable attributes. In addition, the high production cost of biofuels often hinders its commercialisation. Thus, the designed solvent-oil blend needs to achieve both fuel functionality and economic targets to be competitive with the conventional diesel fuel.
In this thesis, a multi-stage computer-aided molecular design (CAMD) framework was employed for bio-oil solvent design. In the design problem, molecular signature descriptors were applied to accommodate different classes of property prediction models. However, the complexity of the CAMD problem increases as the height of signature increases due to the combinatorial nature of higher order signature. Thus, a consistency rule was developed reduce the size of the CAMD problem. The CAMD problem was then further extended to address the economic aspects via fuzzy multi-objective optimisation approach.
Next, a rough-set based machine learning (RSML) model has been proposed to correlate the feedstock characterisation and pyrolysis condition with the pyrolysis bio-oil properties by generating decision rules. The generated decision rules were analysed from a scientific standpoint to identify the underlying patterns, while ensuring the rules were logical. The decision rules generated can be used to select optimal feedstock composition and pyrolysis condition to produce pyrolysis bio-oil of targeted fuel properties.
Next, the results obtained from the computational approaches were verified through experimental study. The generated pyrolysis bio-oils were blended with the identified solvents at various mixing ratio. In addition, emulsification of the solvent-oil blend in diesel was also conducted with the help of surfactants. Lastly, potential extensions and prospective work for this study have been discuss in the later part of this thesis. To conclude, this thesis presented the combination of computational and experimental approaches in upgrading the fuel properties of pyrolysis bio-oil. As a result, high quality biofuel can be generated as a cleaner burning replacement for conventional diesel fuel. |
| first_indexed | 2025-11-14T20:56:07Z |
| format | Thesis (University of Nottingham only) |
| id | nottingham-72215 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:56:07Z |
| publishDate | 2023 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-722152023-02-18T04:40:12Z https://eprints.nottingham.ac.uk/72215/ Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality Chong, Jia Wen As one of the world’s largest palm oil producers, Malaysia encountered a major disposal problem as vast amount of oil palm biomass wastes are produced. To overcome this problem, these biomass wastes can be liquefied into biofuel with fast pyrolysis technology. However, further upgradation of fast pyrolysis bio-oil via direct solvent addition was required to overcome it’s undesirable attributes. In addition, the high production cost of biofuels often hinders its commercialisation. Thus, the designed solvent-oil blend needs to achieve both fuel functionality and economic targets to be competitive with the conventional diesel fuel. In this thesis, a multi-stage computer-aided molecular design (CAMD) framework was employed for bio-oil solvent design. In the design problem, molecular signature descriptors were applied to accommodate different classes of property prediction models. However, the complexity of the CAMD problem increases as the height of signature increases due to the combinatorial nature of higher order signature. Thus, a consistency rule was developed reduce the size of the CAMD problem. The CAMD problem was then further extended to address the economic aspects via fuzzy multi-objective optimisation approach. Next, a rough-set based machine learning (RSML) model has been proposed to correlate the feedstock characterisation and pyrolysis condition with the pyrolysis bio-oil properties by generating decision rules. The generated decision rules were analysed from a scientific standpoint to identify the underlying patterns, while ensuring the rules were logical. The decision rules generated can be used to select optimal feedstock composition and pyrolysis condition to produce pyrolysis bio-oil of targeted fuel properties. Next, the results obtained from the computational approaches were verified through experimental study. The generated pyrolysis bio-oils were blended with the identified solvents at various mixing ratio. In addition, emulsification of the solvent-oil blend in diesel was also conducted with the help of surfactants. Lastly, potential extensions and prospective work for this study have been discuss in the later part of this thesis. To conclude, this thesis presented the combination of computational and experimental approaches in upgrading the fuel properties of pyrolysis bio-oil. As a result, high quality biofuel can be generated as a cleaner burning replacement for conventional diesel fuel. 2023-02-18 Thesis (University of Nottingham only) NonPeerReviewed application/pdf en cc_by https://eprints.nottingham.ac.uk/72215/1/PhD%20Thesis_Chong%20Jia%20Wen.pdf Chong, Jia Wen (2023) Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality. PhD thesis, University of Nottingham. computer-aided molecular design pyrolysis bio-oil biofuel rough set machine learning solvent design |
| spellingShingle | computer-aided molecular design pyrolysis bio-oil biofuel rough set machine learning solvent design Chong, Jia Wen Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality |
| title | Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality |
| title_full | Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality |
| title_fullStr | Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality |
| title_full_unstemmed | Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality |
| title_short | Computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality |
| title_sort | computational and experimental studies on the reaction mechanism of bio-oil components with additives for increased stability and fuel quality |
| topic | computer-aided molecular design pyrolysis bio-oil biofuel rough set machine learning solvent design |
| url | https://eprints.nottingham.ac.uk/72215/ |