Systematic approaches to design chitosan functionalised 3D-printed monolith with AI technology for colour removal in wastewater treatmentTools Mohd Yusoff, Nurul Husna (2024) Systematic approaches to design chitosan functionalised 3D-printed monolith with AI technology for colour removal in wastewater treatment. PhD thesis, University of Nottingham.
AbstractAccess to clean water is a fundamental problem that has been emphasised by the United Nations and World Health Organization. Water is one of the vital necessities for humans to stay alive and prevent disease. However, due to the increasing world population and climate change, it is predicted water resources will be stressed and limited due to poor quality. Various water treatment processes have been developed based on the presence of contaminants in water to solve the urgent issue. To date, 3D printing has been utilised over the conventional method in water treatment due to its capability to manufacture geometrically complex structures and cost-effectiveness with minimum wastage of materials. The latest approaches of synergising additive manufacturing (AM) and surface functionalisation have drawn tremendous attention in recent years due to their outstanding performance to achieve the requirements of commercial applications. In this work, a critical review on 3D printable functional materials in water application was prepared. The robust approach of applying surface functionalisation strategies for additively manufactured material have been discussed along with emerging application water application. This project takes the approach of treating textile wastewater through experimental work with enhancement strategy and modelling using artificial intelligence. The first experimental data that focusing on the development of novel design on chitosan – grafted 3D -printed monolith were established. A set of 81 experimental data were run to evaluate the adsorption performance of the chitosan-grafted monolith with series of characterisation analysis was studied. It was reported the removal efficiency (R%) of methyl orange (MO) dye ranged from 20.8 to 90.4% and the equilibrium uptake capacity (K) was ranged from 1 to 12.62 (mg/g) after 2 h. The percent reduction for BOD and COD are 62.4% and 2 - 3.9%, respectively when using real textile wastewater. The recyclability of the chitosan - grafted PEGDA monolith was verified by evaluating its adsorption for four consecutive cycles and retained at 78%. The combined methodology of integrating 3D printing technique with surface grafting techniques has effectively led to the creation of an innovative environmental-friendly wastewater treatment technology. This approach embodies principles of sustainability and zero waste by eliminating the necessity for replacing adsorbents and removing sludge. This approach also is a cost-effective solution as it can reduce the cost of spending on the chemicals and maintenance. Moreover, structural design of body centred cubic (BDD) has the best adsorption performance with high surface area and permeability which further enhance contact point between the adsorbent with the adsorbate. Additionally, the resin formulation of PEGDA and addition of 3wt.% of HDDA has shown to be the best candidate to be developed as green water technology, whereas silane - grafted with GLYMO on the 3D-printed monolithic structure has achieved a good hydrophobicity, reduction in moisture absorption content, good mechanical as well as chemical stability. This approach has extended the functionality of the 3D-printed monolith for utilisation in aquatic settings in contrast to conventional polymeric materials that exhibit a propensity for water absorption. Further to this, the results of the convective hot air drying have better performance while retaining the surface morphology and removed adequate moisture content within 6% for better storage and lifespan of the functional groups of the chitosan to be used as an adsorbent. Meanwhile, the electron beam treatment with 200cGy has successfully increased the adsorption performance to 96.6% within 80 min. The integration of four stages of the enhancement strategy is anticipated to facilitate the development of a 3D-printed enhanced adsorbent that exhibits excellent adsorption efficacy, robust mechanical and chemical durability, and shorter adsorption time for the treatment of textile wastewater. Lastly, the prediction of water analysis using convolutional neural network and AlexNet was performed to predict the colour of textile wastewater based on real-time images. It was found that both networks demonstrated excellent results in terms of its ability to accurately classify data, as well as its efficient training and testing time. For large data sets, CNN has the highest training accuracy with 94.2%, while AlexNet has better performance in handling small date sets with 89.8%. The conclusions and future works are included at the end of this thesis.
Actions (Archive Staff Only)
|
Tools
Tools
![[img]](/style/images/fileicons/application_pdf.png)
