Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution
Dissertation (MEng (Chemical Engineering))--University of Pretoria, 2026.
Saved in:
| Other Authors: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
University of Pretoria
2026
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1869483824662970368 |
|---|---|
| access_status_str | Open Access |
| author2 | Iwarere, Samuel Ayodele |
| author_browse | Iwarere, Samuel Ayodele |
| author_facet | Iwarere, Samuel Ayodele |
| collection | Thesis |
| dc_rights_str_mv | © 2024 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. |
| description | Dissertation (MEng (Chemical Engineering))--University of Pretoria, 2026. |
| format | Thesis |
| id | oai:repository.up.ac.za:2263/107953 |
| institution | University of Pretoria (South Africa) |
| language | English |
| last_indexed | 2026-07-01T04:05:08.518Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UPSpace — University of Pretoria Institutional Repository |
| publishDate | 2026 |
| publishDateRange | 2026 |
| publishDateSort | 2026 |
| publisher | University of Pretoria |
| publisherStr | University of Pretoria |
| record_format | dspace |
| source_str | UPSpace — University of Pretoria Institutional Repository |
| spelling | oai:repository.up.ac.za:2263/107953 Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution Iwarere, Samuel Ayodele leratotau24@gmail.com Merckel, Ryan David Tau, Lerato B.A. UCTD Sustainable Development Goals (SDGs) Thermogravimetric analysis Lignocellulose biomass Kraft pulp Pyrolysis Characterisation Dissertation (MEng (Chemical Engineering))--University of Pretoria, 2026. The accurate determination of hemicellulose, cellulose, and lignin in wood and pulp materials is a critical step in the paper and pulp industry as it provides insight into the pulping mills efficiencyand informs product development. Traditionally, such analyses rely on laborious, chemical-intensive methods that are time-consuming, costly, and require hazardous reagents. This study investigates a rapid, computationally driven thermal method for lignocellulose characterisation using thermogravimetric analysis (TGA) coupled with the independent parallel reaction (IPR) kinetic model. The IPR model was selected for its computational efficiency and ability to deconvolute overlapping peaks in the derivative thermogravimetric (dTG) curves, which show the rate of mass loss as temperature increases. The initial kinetic parameters, specifically activation energy (Ea) and pre-exponential factor (A), for each component were estimated using a novel top-and-tail approach which linearises selected regions of the dTG curve. The study workflow comprised of three levels: model verification, validation, and sensitivity analysis. Model verification was performed using five published pinewood datasets. The model achieved low root-mean-square error (RMSE) values ranging from 0.015 to 0.100, high adjusted R² values (97.74 % – 99.96 %), and fit qualities above 97.00 %, while the deviations in predicted component mass fractions compared to literature values ranged from 0.2 % to 8.4 %. This verified the model’s ability to accurately deconvolute TGA data and predict lignocellulose composition for pine wood. Subsequently, model validation was conducted using industrially sourced pine and Eucalyptus wood, and their derived pulp samples. The model successfully predicted the hemicellulose, cellulose and lignin content of the wood samples, achieving RMSE values of 0.063 for pine and 0.060 for the Eucalyptus. However, predictions for kraft pulp samples revealed systematic overestimation of lignin content, suggesting that the model encounters limitations in handling samples with low lignin fractions and highlights the need to account for component interactions. This peak-focused deconvolution model demonstrated rapid characterisation by converging within an hour of optimisation, thus, confirming the method’s speed in determining lignocellulose content in woody biomass. Synthetic mixtures were formulated from known quantities of xylan, microcrystalline cellulose, and kraft lignin to represent typical lignocellulosic compositions of wood and pulp. Application of the model to these mixtures showed reasonable predictions for the wood-like compositions (Mixes 1–4), with deviations ranging from 1.7 % to 12.6 %, while mixtures resembling pulp samples (Mixes 5–6) exhibited substantial lignin overestimation. These findings corroborate the validation results, indicating that the model performs well for complex, lignin-rich mixtures but requires refinement for low-lignin materials. Potential refinements for future work include incorporating hybrid kinetic deconvolution approaches to better capture component interactions, supporting model predictions through traditional wet-chemistry methods. Finally, sensitivity analysis was conducted by varying heating rates (5 °C/min, 10 °C/min, 20 °C/min), performing local parameter perturbations (±20%), and altering initial parameter estimates. The model proved sensitive to heating rate variations as the model predictions varied drastically at a slow heating rate (5 °C/min) and a faster heating rate (20 °C/min), the most reliable and accurate model predictions were determined at 10 °C/min. Local sensitivity analysis showed that the kinetic parameters for hemicellulose and cellulose had the largest influence on composition predictions. It was also discovered that parameter initialisation influenced model convergence as some optimisation processes reached local minima instead of the global minima when the initial estimates were varied randomly. These results indicate that the number of model parameters could be efficiently reduced to mitigate parameter compensation effects, and that optimisation across multiple heating rates should be explored in future work to improve model robustness. Overall, this study demonstrates the potential of TGA combined with the IPR model as a rapid, reagent-free, and computationally efficient method for lignocellulosic characterisation in woody biomass. While improvements are needed to refine predictions for pulp samples, the approach provides a promising tool for frequent compositional analysis, reducing reliance on conventional chemical-intensive techniques and supporting industrial process monitoring. Paper Manufacturers Association of South Africa (PAMSA) Chemical Engineering MEng (Chemical Engineering) Unrestricted Faculty of Engineering, Built Environment and Information Technology SDG-09: Industry, innovation and infrastructure SDG-12: Responsible consumption and production 2026-02-09T07:31:25Z 2026-02-09T07:31:25Z 2026-04 2026 Dissertation * A2026 http://hdl.handle.net/2263/107953 10.25403/UPresearchdata.31268203 en © 2024 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. application/pdf University of Pretoria |
| spellingShingle | UCTD Sustainable Development Goals (SDGs) Thermogravimetric analysis Lignocellulose biomass Kraft pulp Pyrolysis Characterisation Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution |
| title | Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution |
| title_full | Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution |
| title_fullStr | Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution |
| title_full_unstemmed | Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution |
| title_short | Rapid component-specific characterisation of woody biomass via peak-focused thermogravimetric deconvolution |
| title_sort | rapid component specific characterisation of woody biomass via peak focused thermogravimetric deconvolution |
| topic | UCTD Sustainable Development Goals (SDGs) Thermogravimetric analysis Lignocellulose biomass Kraft pulp Pyrolysis Characterisation |
| url | http://hdl.handle.net/2263/107953 |