Research Projects in 2025-2026

Ultrasound-Microwave Assisted Innovative Oat Hull Valorisation of Arabinoxylan and Structure Effects (PhD)

N Lin, D Aggarwal, S Hannon, S Ojha, Da-Wen Sun and B K Tiwari

Sponsors: CSC-UCD Scholarship Scheme

The study explored the efficacy of different green extraction technologies, including high-intensity ultrasound (US-High), high-intensity microwave (MW-High), and combined ultrasound–microwave (US–MW-High), in extracting arabinoxylan (AX) from oat hulls, using room-temperature water extraction as the control. MW-High showed the highest AX content (36.84 ± 1.37 g/100 g), alongside the highest arabinose (10.20 ± 0.00 g/100 g) and xylose (19.39 ± 0.36 g/100 g), greatly exceeding the control (9.69 ± 0.00 g/100 g AX). By contrast, US-High and US–MW-High yielded lower AX contents (6.46 ± 0.84 and 4.55 ± 0.57 g/100 g, respectively). The results also indicated that US–MW-High produced the highest β-glucan level (2.89 ± 0.02%), whereas MW-High showed a moderate β-glucan content (1.25 ± 0.04%) and US-High remained low (0.08 ± 0.00%), highlighting clear differences in co-extracted polysaccharides among extraction routes.

---

Ultrasound-Assisted Enzymatic Extraction and Optimization of Protein Recovery from Sunflower Oilseed Meal (PhD)

Y C Ban, Da-Wen Sun and B K Tiwari

Sponsors: CSC-UCD Scholarship Scheme

This study evaluates the differences and potential synergy between ultrasound (US) and enzymatic treatments (Viscozyme, Cellulase, and α-Amylase) to improve protein extraction from sunflower oilseed meal (SOM). The US treatment achieved the highest protein recovery (55.44%), but with a much lower protein purity (47.68%), suggesting ultrasound may cause structural disruption and/or shift proteins into the soluble fraction. In contrast, the enzyme-only group achieved the highest protein purity (78%), indicating that enzymes help preserve protein quality and purity, although the gains in recovery were more limited (maximum 50.12%). To optimize the enzymatic approach, a mixture design was used to quantify the contributions and interactions of the three enzymes and to develop a predictive model for Protein B recovery: Protein B-Recovery (%) = 48.50 × Viscozyme (%) + 36.83 × Cellulase (%) + 46.52 × α-Amylase (%). Based on the model prediction and optimization, the current optimal enzyme formulation is Viscozyme 0.237% + α-Amylase 1.763% (w/w).

---

Continuous Monitoring of Moisture Loss of Beef, Beetroot and Banana Slices During Microwave Vacuum Dehydration by using THz-TDS combined with Transformer-Based Neural Network (PhD)

Y Fu, Z H Zhang and Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Microwave vacuum dehydration (MVD) has emerged as a preferred alternative to conventional methods such as hot air drying, offering faster dehydration rates while better preserving product quality. Despite these advantages, challenges remain in implementing effective real-time monitoring systems and accurate dehydration prediction methods during the MVD process. This study investigated the feasibility of using terahertz time-domain spectroscopy (THz-TDS) to continuously monitor the drying kinetics of beef, beetroot, and banana slices during MVD without interrupting the dehydration process. Polytetrafluoroethylene (PTFE) was demonstrated as the most suitable airhose material among polyethene (PE), PTFE, and quartz, with the highest transmittance of 0.824. Using the deep learning model of a transformer-based neural network (TbNN) introduces the self-attention mechanisms to extract features at characteristic frequencies. It successfully correlated THz-TDS transmittance data with actual moisture loss of samples with a prediction accuracy of 0.96, which shows excellent generalisation capability of this TbNN model on such a small dataset size.  Besides, the calibration strategy successfully improves the accuracy from 0.94 to 0.96, with a regression coefficient of R=0.98328. The integration of these sensing and analytical techniques offers a valuable framework for improving industrial processing control while broadening the applications of THz-TDS technology across agricultural and food production sectors.

---

Advancing Fruit Quality Detection through Artificial Intelligence (PhD)

Y Fu and Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Fruit quality monitoring has become increasingly critical in modern agriculture and postharvest management due to its direct impact on food security, economic sustainability, and consumer health. Traditional methods involving destructive sampling methods and manual inspection are often labour-intensive and time-consuming. The emergence of non-destructive detection methods and artificial intelligence (AI) enables transformative potential by enabling rapid and high-throughput quality assessment. This review systematically reviews the applications of integrating AI with advanced sensing technologies to address these challenges and enhance precision in fruit quality evaluation from aspects of pre- and post-harvest stages.

This review analyses applications of spectroscopic techniques (e.g., NIR, THz), imaging methods (e.g., hyperspectral, X-ray), and sensor-based systems (e.g., electronic nose) in the fruit quality detection. It evaluates the application of AI, including traditional machine learning (e.g., SVM, RF), deep learning (e.g., CNNs, GANs), and hybrid models, across pre- and postharvest stages. Special attention is given to real-time implementations, edge AI, and multi-modal data fusion to optimise accuracy and efficiency. The study synthesises advancements in ripeness assessment, defect detection, microbial monitoring, and automated sorting systems.

AI-driven approaches demonstrate remarkable success, achieving over 95% accuracy in tasks such as ripeness classification and bruise detection, while reducing postharvest losses by 20–40%. However, challenges persist, including model brittleness in field conditions, high costs of advanced sensors, and the need for standardised protocols. Future directions emphasise explainable AI (XAI), IoT-enabled digital twins, and autonomous agri-robotics to bridge the gap between research and industry adoption. This review reveals the transformative potential of AI in fruit quality monitoring, provided that interdisciplinary collaboration and scalable solutions are prioritised.

---

Valorization of Dairy By-Products Acid Whey Powder by Schizochytrium Sp. S31: Towards an Alternative Eco-Pathway for Docosahexaenoic Acid Production (PhD)

M S Xiao, Da-Wen Sun and R Halim

Sponsors: CSC-UCD Scholarship Scheme

This study employed the heterotrophic marine microalga Schizochytrium sp. S31 cultivated on acid whey powder (AWP) as the primary carbon source to explore the potential for algal DHA production. Batch cultivation experiments demonstrated that a salinity of 50% seawater (17.5 g/L) provided optimal conditions for the efficient utilization of lactose-derived sugars, supporting robust biomass production of up to 16 g/L and enhanced lipid biosynthesis. High total fatty acid content was achieved, ranging from 39.5% to 78.1% of biomass, with DHA accounting for approximately 10% of the total biomass. Fatty acid profiling revealed that excess carbon availability promoted lipid accumulation following nitrogen depletion, primarily through increased synthesis of C14 and C16 fatty acids. These results indicate that AWP can effectively replace purified carbon sources without compromising DHA yield or quality. This valorization strategy integrates dairy by-product management with microbial omega-3 production, offering an alternative eco-pathway that enhances resource efficiency and supports circular bioeconomy principles. Overall, this study highlights the potential of Schizochytrium sp. S31 cultivated on acid whey powder as a sustainable feedstock for DHA production.

---

Comparative Evaluation of Salmon By-Products Oil Bio-Refinery by Ultrasound and Microwave Technologies: Fatty Acid Composition and Nutritional Quality Indices (PhD)

W R Dong, Da-Wen Sun and B K Tiwari 

Sponsors: CSC-UCD Scholarship Scheme

This study evaluated thermal (wet rendering, microwave) and non-thermal (cold pressing, ultrasound) extraction technologies, exploring the synergistic potential of ultrasonic-microwave hybrid technology for salmon by-product oil recovery. We quantified oil recovery and emulsification under different processing conditions, analysed fatty acid profiles, and calculated key nutritional indices. Results indicate that microwave-assisted extraction significantly promotes the extraction efficiency and reduces thermal stress. However, ultrasonication-induced emulsification negatively impacts oil separation, and potential oxidation slightly diminishes key nutritional indicators. This underscores the necessity for process optimization. These findings provide foundational insights for efficient, environmentally friendly utilization of salmon by-products, contributing to sustainable marine resource management and advancing circular economy practices.

---

Comparative Studies on Shrimp Chitin Valorisation by Various Ultrasonic Irradiation and Characterization (PhD)

W R Dong, Da-Wen Sun and B K Tiwari 

Sponsors: CSC-UCD Scholarship Scheme

This work establishes a systematic framework for chitin recovery from shrimp shells using ultrasound-assisted processing as an alternative to conventional chemical treatment. Multiple ultrasound configurations, including low-frequency probe ultrasound (USP), high-frequency ultrasound baths (USB), and an ultrasound–microwave hybrid system (US-MW), were integrated into standardized deproteinization and demineralization steps. By applying identical chemical conditions and short treatment times, the study enables a direct comparison of energy delivery modes and dominant mechanisms, such as cavitation, thermal effects, and mass transfer enhancement. The approach highlights how different ultrasound setups influence the balance between mechanical disruption and thermally assisted reactions during chitin extraction. Importantly, the workflow demonstrates the feasibility of substantially reducing processing time while maintaining effective component separation. 

---

Comparative Studies on Heterogeneous Deacetylation under Various Ultrasonic Irradiations and Characterization (PhD) 

W R Dong, Da-Wen Sun and B K Tiwari 

Sponsors: CSC-UCD Scholarship Scheme This study innovatively explores various ultrasound systems as alternative strategies to achieve rapid and efficient chitin deacetylation. 

Different systems were systematically compared under standardized 15-minute treatments: low-frequency ultrasound probe (USP), high-frequency ultrasound plate (USB), ultrasound–microwave combination (US-MW). Results showed that USB and US-MW achieved effective deacetylation comparable to that of traditional 3-hour high-temperature treatments. The USB method obtained chitosan with great yield and well-preserved morphology, while US-MW facilitated highly efficient transformation with structural depolymerization. In contrast, USP exhibited limited performance due to insufficient thermal and cavitation effects. Structural and functional characteristics of the chitosan analogs were confirmed through FTIR, XRD, SEM, TGA-DSC, and HPLA-SEC. This work provides foundational insights into scalable, green, and energy-efficient chitosan production, highlighting the potential of high-frequency plate and ultrasound–microwave hybrid technologies in sustainable biopolymer manufacturing. 

---

Valorisation of Haddock Side Streams by Enhancing Protein Recovery Using Ultrasound (US) and High-Pressure Processing (HPP) and Assessing Its in Vitro Digestibility (PhD) 

W R Dong, Da-Wen Sun and B K Tiwari 

Sponsors: CSC-UCD Scholarship Scheme

This study innovatively explores the combination of ultrasound (US) and high-pressure processing (HPP) with isoelectric solubilisation precipitation and Alcalase hydrolysis, which offers sustainable ways to improve protein recovery from fish waste. The findings indicated that alkali extraction with 50% US amplitude and 200 MPa pressure treatment yielded a maximum at 70.49 ± 0.01%, which also aligns with optimal digestibility and antioxidant properties. During the processing, the HPP and US combination displayed a synergy. The implications for the food industry are significant, offering a path towards more sustainable and efficient protein extraction techniques that can mitigate environmental impacts and contribute to circular economy principles.  

---

Comparative Study of Machine Learning Models for Quantitative Analysis of Alaria esculenta Extract Samples Using THz Imaging (PhD)

Q X Li, Da-Wen Sun

Sponsors: CSC-UCD Scholarship Scheme

Alginate purity is an important chemical indicator for evaluating the quality and extraction efficiency of marine biological polysaccharides. In this study, a Terahertz (THz) spectral imaging system was applied for the first time to the quantitative prediction of alginate purity in seaweed extract samples. To evaluate the optimal modelling strategy, three machine learning algorithms were developed and compared, including Partial Least Squares Regression (PLSR), Gaussian Process Regression (GPR), and a three-dimensional convolutional neural network (3D-CNN). The 3D-CNN was built to capture the combined spatial-spectral features of the THz data, while PLSR and GPR were based on mean spectral features. The results showed that the GPR model achieved the highest prediction performance with an  of 0.9162, the 3D-CNN model achieved a reliable prediction performance with an  of 0.8414. It provides a new possibility for the non-destructive quantitative evaluation of biopolymer purity in food and pharmaceutical materials.

---

Multifactorial Analysis of Allergen Variability in Alaria Esculenta Seaweed: Influence of Drying Techniques, Harvesting Methods, and Pre-Treatments (PhD)

G Y Dong, Da-Wen Sun and B. K. Tiwari

Sponsors: CSC-UCD Scholarship Scheme

This study investigates the effects of different harvesting methods, drying techniques, and pre-treatments on allergen management in Alaria esculenta (an Irish brown seaweed). The allergens hel as 1, pen i 1 (tropomyosin proteins), and parvalbumin (PV) were examined in aquaculture and wild-grown samples subjected to oven drying (OD), freeze-drying (FD), and infrared drying (IRD). Cultivated samples also received pre-treatments, including hot water blanching (HWB), ultrasound (US), and microwave (MW) blanching, prior to drying. Allergen reactivity was quantified via ELISA to provide precise allergen measurements. Results showed that all three drying methods significantly reduced allergen levels in cultivated samples, with OD being the most effective approach, while FD demonstrated comparatively lower efficiency. Importantly, in wild-grown samples, hel as 1 level unexpectedly increased following drying, suggesting a distinct environmental resilience mechanism that merits further investigation. Among pre-treatments, ultrasound blanching proved the most effective, whereas microwave blanching, particularly at lower power, yielded limited allergen reduction. Findings indicate that tailored drying processes are critical for allergen mitigation in seaweed, with heat-induced drying methods and ultrasound identified as promising strategies for effective allergen management in Alaria esculenta.

---

Reducing Anti-Nutritional Factors in Pea Protein Using Advanced Hydrodynamic Cavitation, Ultrasonication, and High-Pressure Processing Technologies (PhD)

G Y Dong, Da-Wen Sun and B. K. Tiwari

Sponsors: CSC-UCD Scholarship Scheme

The study assessed the impact of advanced technologies — high-pressure processing (HPP), ultrasonication (US), and hydrodynamic cavitation (HDC) — during pea protein extraction on anti-nutritional factors (ANFs) in pea protein isolates (PPI) and the residues. US (20 kHz) and HDC (50 passes) were used for protein extraction, while HPP (200, 300, 400 MPa) served as a pre-treatment before traditional stirring-based extraction. ANF levels in the PPIs from these methods were compared to those from conventional stirring-based extraction (CE) and a commercial pea protein isolate (Cm). Trypsin inhibitor activity (TIA) and tannin levels decreased across methods, with HDC showing the highest TIA reduction (66.09 %) and CE reducing tannin content the most (75.59 %). HPP (200 MPa) pre-treatment combined with stirring led to moderate ANF reductions but maximum protein recovery (72.74 %). However, phytic acid levels increased in all PPIs. Overall, HDC and HPP pre-treatment demonstrated potential in enhancing protein recovery and reducing specific ANFs in PPI.

---

Novel Methods for Extracting Protein from Peeled Potato (PhD)

Z P Hu, G Y Dong, X L Zhu, Da-Wen Sun and B K Tiwari

Sponsors: CSC-UCD Scholarship Scheme

This study investigates the efficiency, purity, and energy consumption of various protein extraction methods from peeled potato, comparing conventional techniques with enzyme-assisted and ultrasound-assisted extraction (UAE) under water and alkaline conditions. Results indicated that alkaline-based UAE was the most effective strategy, achieving the highest yield (2.99 g/100g)—a two-fold increase over conventional water extraction—and the most optimal energy efficiency (3.57 kWh/100g pure protein). FTIR and structural analysis revealed that ultrasonic and alkaline treatments induced protein unfolding, characterized by a significant decrease in β-sheet structures and an increase in random coils, which potentially enhances solubility. Additionally, amino acid profiling demonstrated a relative increase in essential amino acids in the extracted products, confirming the potential of alkaline-based UAE as a sustainable strategy for producing high-quality nutritional protein.

---

Novel Methods for Extracting Protein from Whole Potato (PhD)

Z P Hu, G Y Dong, Da-Wen Sun and B K Tiwari

Sponsors: CSC-UCD Scholarship Scheme

This study aimed to simulate the valorisation of oversized potatoes found in industrial side-streams by utilizing whole potatoes as the raw material for protein extraction. The research evaluated the efficiency of various emerging technologies, including ultrasound, high pressure, microwave, and combined microwave-ultrasound treatments, on both fresh and dried potato samples. Results indicated that the combined microwave and ultrasound treatment yielded the highest protein recovery rates, achieving 1.4 g/100g for dried samples and 1.37 g/100g for fresh samples. The extracted proteins were subsequently subjected to Fourier Transform Infrared Spectroscopy (FTIR) and Scanning Electron Microscopy (SEM) to analyse structural integrity, alongside heavy metal content analysis to ensure product safety and quality.

---

Ultrasound-Assisted Emulsions Stabilized by Lentil Protein Fibril Systems: Structural, Physicochemical, and Stability Characteristics (PhD)

S Y Wang, Da-Wen Sun and M Song

Sponsors: CSC-UCD Scholarship Scheme

Ultrasound-assisted emulsification is a high-energy technique that has demonstrated efficacy in reducing droplet size, narrowing size distribution, and enhancing emulsion stability. Despite these promising attributes, little is known about the effect of ultrasound treatment on emulsions stabilized by lentil protein fibrils, particularly under varying pH conditions. The objective of the study was to investigate the impact of ultrasound at varying sonication times on the structure and stability of oil-in-water emulsions stabilized by LPI fibril systems prepared at different pH conditions. Emulsion characteristics, such as creaming stability, droplet size distribution, microstructure, and rheological properties, were analyzed, along with evaluations of thermal and long-term stability. Emulsions prepared at pH 2 exhibited higher surface hydrophobicity and absolute ζ-potential, smaller particle size, and greater interfacial properties compared to those at pH 7, contributing to superior emulsion stability. Among all treatments, a 10 min ultrasound treatment at pH 2 produced emulsions with optimal creaming, storage, and thermal stability. At pH 7, ultrasound treatment led to limited improvements in emulsion stability, particularly with extended treatment durations. Nevertheless, the influence of ultrasound was more significant under acidic conditions.

---

Inactivation Efficacies of Plasma-Activated Water (PAW) Against E. coli Biofilms Formed on Fresh Produce and Food Contact Surfaces (PhD)

Y L Zhao, Da-Wen Sun and B K Tiwari

Sponsors: CSC-UCD Scholarship Scheme

The study evaluated the inactivation efficacy of plasma-activated water (PAW) in removing E.coli biofilm formed on fresh produce (spinach) and food packaging material (biodegradable material). Post-PAW viability was assessed by plate counts, and cell sublethal injury was also quantified. The efficacy of PAW was increased with increasing plasma generation time and PAW exposure time for biofilm formed on both surface types. Furthermore, PAW treatment significatly reduce the biomass of E.coli biofilm formed on spinach after 15 min of exposure. After 24h of storage at 4 °C, no significant regrowth of bacterial colonies was found on spinach. These findings are of significance to the food industry for the development of effective methods for food and food contact surface decontamination using PAW.