Crystallization fouling and liquid retention on engineered surfaces during liquid-vapor phase change

Crystallization fouling is the accumulation of undesired inorganic salts on surfaces, which occurs in numerous natural and industrial scenarios, such as desalination, food processing, and power generation. In most applications, especially there is heat transfer involved, the formation of scale layer with low thermal conductivity and porous structure drastically deteriorates system efficiency by degrading heat transfer and restricting flow channel. Regular cleaning and maintenance of heat exchangers are required due to excessive foulant accumulation with additional cost and system downtime. Therefore, mitigation of crystallization fouling on heat transfer surface is crucial to improve long-time performance and reduce energy consumption. Crystallization fouling is investigated in experimental study and numerical modeling at multiple scales to identify potential approach to decrease scale formation and promote crystal removal. Fouling experiments are conducted in tank reactor, evaporating droplet, and falling film evaporator to mimic different thermal and hydraulic conditions. Various materials and engineered microstructures have been tested, including stainless steel with different roughness and microgrooves, PEEK, 3D printed polymer, and superhydrophobic coating. Crystal morphology, deposition mass, and surface wettability are compared to evaluate the anti-fouling performance. It is found that crystal size, distance, and contact area with substrate undergo a transition with surface roughness and wettability, which results in different porosities, effective thermal conductivities, and adhesion of fouling layer. The observation of crystallization in evaporating droplet reveals that surface energy, microstructures, and ion species alter crystal formation and droplet pining/depinning, due to the complex coupling of concentration and temperature profiles with evaporation. In falling film evaporation, localized fouling is achieved with wettability gradient. In comparison with plain metallic tubes, heat transfer deterioration with fouling is delayed by applying hybrid wettability pattern. Numerical modeling has been established to predict crystallization fouling process during seawater falling-film flow over a horizontal tube bundle, which has the scaling estimated based on the deposition theory with the heat transfer process coupled including the in-tube steam condensation, conduction through the tube wall and the scale layer, and falling-film evaporation. The spatial and temporal variations of temperature, heat transfer coefficient, and scale thickness have been predicted, and the effects of variations in process parameters of steam, feed seawater, and tube properties on the scale thickness, evaporation rate, and heat transfer coefficient for falling-film evaporation are analyzed. In order to identify optimal conditions, batch numerical simulations of heat and mass transfer are conducted to create a database, with which machine learning and multi-objective optimization algorithm are performed. The deep neural network serves as surrogate model to efficiently generate function evaluations for optimization. Non-dominated sorting genetic algorithm is then introduced to determine and analyze the optimal pareto front for two and three objectives in desalination criteria. Tradeoffs between mitigating scale formation and enhancing desalination performance are evaluated in optimizations for different objectives. Potential optima are identified and can be applied as guidelines to determine evaporator design and operating conditions.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2022

Author dc:creator

• Jin, Hongqing

Contributors dc:contributor

• Wang, Sophie
• Smith, Kyle
• Wang, Xinlei
• Miljkovic, Nenad

• crystallization fouling
• thermal desalination
• falling film evaporation
• numerical simulation
• machine learning
• optimization

Statement dc:rights

• Copyright 2022 Hong-Qing Jin

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/115423

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F115423