Experimental and modelling of emulsified oil adsorption using functionalized magnetic nanoparticles

Hamedi, Hamideh (2024) Experimental and modelling of emulsified oil adsorption using functionalized magnetic nanoparticles. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[img] [English] PDF - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (13MB)


Conventional oil-water separation techniques have limited effectiveness in separating emulsified oil separation from wastewater due to the high stability of oil-in-water (o/w) emulsions. Magnetic nanoparticles (MNPs) application for emulsified oil separation from wastewater is becoming more popular, despite their inherent instability due to their high chemical activity. Thereby, their tendency to agglomerate, precipitate, and oxidize by air, results in decreasing magnetism and dispersibility. Stabilization of MNPs is needed by generating a protective coating layer on their surfaces to prevent their agglomeration and enable MNPs to disperse properly in the aqueous content through changing their functionality. Functionalizing MNPs using amphiphilic compounds, featuring both hydrophilic and oleophilic properties, improves their dispersivity, which is required for an efficient demulsification. This research aims to study demulsification performance of functionalized MNPs using surfactants, as amphiphilic compounds, for capturing emulsified oil droplets from nanoemulsions. To this end, different sizes of Fe3O4 nanoparticles are functionalized using sodium dodecyl sulfate (SDS) as an anionic surfactant, and cetyltrimethylammonium bromide (CTAB) as a cationic surfactant. The functionalized particles are characterized using Transmission Electron Microscopy (TEM), Scanning Electron Microscopy (SEM), Energy Dispersive X-ray (EDX), contact angle (CA), hydrodynamic diameter, and zeta potential measurement. We study the effect of size and concentration of MNPs, coating materials, and surfactant to MNPs mass ratio on demulsification performance to find the optimum MNP feature with the highest oil separation efficiency (SE), which is measured via Gas Chromatography equipped with Flame Ionization Detector (GC-FID) analysis. The demulsification performance of the functionalized MNPs is tested for oil adsorption from 1000 ppm dodecane-in-water nanoemulsion, containing ultra-small droplets (almost 300 nm). The oil-water separation results reveal the superior performance of 0.5 g/l smaller size CTABcoated particles (MNP-S@CTAB) with lower CTAB to MNP mass ratio of 0.4 (SE = 99.80%), compared to the bare MNPs, and MNP-S@SDS that achieve SE around 57.46% and 86.1%, respectively. This high SE is attributed to (i) the more positive surface charge density on CTABcoated particles compared to bare and SDS-coated ones, which is verified through zeta potential measurements; (ii) more hydrophilicity of the CTAB-coated MNPs (compared to SDS-coated ones) as evidenced by the WCA analysis; and (iii) less aggregation of MNP@CTAB in the aqueous phase as illustrated in the SEM and TEM images. Moreover, better performance of smaller functionalized particles compared to the larger ones is attributed to their higher surface energy and charge density. The reusability test indicates an excellent cycling stability of MNP-S@CTAB after ten cycles. In the modelling phase of the research, to obtain a better understanding of the oil capturing behavior and the rate of adsorption, systematic investigations about isotherm and kinetic models are conducted. To this end, we employ three adsorption isotherms, including Langmuir, Freundlich, Temkin, and three kinetic models, including pseudo-first-order (PFO), pseudo-secondorder (PSO), and intra-particle diffusion (IPD). The oil adsorption equilibrium data is estimated by oil adsorption capacity measurements using GC-FID. It is found that the Freundlich isotherm and PFO kinetic models are the best fit to the experimental equilibrium data, verifying a multilayer heterogeneous physical adsorption of oil onto MNPs. Moreover, optimization of the oil adsorption process using the functionalized MNPs is investigated to achieve a higher oil adsorption capacity. To this end, smart models based on artificial intelligence (AI) strategies such as least squares support vector machines (LSSVM) hybridized with the coupled simulated annealing (CSA) algorithm, adaptive network-based fuzzy inference system (ANFIS), and gene expression programming (GEP) are applied to assess the non-linear relationships of effective input and output variables. Oil adsorption capacity is considered as the target variable, and the oil concentration, mixing time, and MNP dosage are selected as the input variables. After conducting experiments, 149 data points are obtained, divided into two parts; 80% for the training process and the remaining 20% for the testing step of modelling. Detailed smart model evaluation and error analysis indicate that the LSSVM-CSA model predicts better than ANFIS and GEP in terms of accuracy, with the highest

Item Type: Thesis (Doctoral (PhD))
URI: http://research.library.mun.ca/id/eprint/16329
Item ID: 16329
Additional Information: Includes bibliographical references
Keywords: oil adsorption, nanoemulsion, magnetic nanoparticle, isotherm and kinetic model, machine learning
Department(s): Engineering and Applied Science, Faculty of
Date: February 2024
Date Type: Submission
Digital Object Identifier (DOI): https://doi.org/10.48336/MKPQ-X374
Library of Congress Subject Heading: Magnetic nanoparticles; Oil separators; Oil-in-water emulsions; Oils and fats--Absorption and adsorption

Actions (login required)

View Item View Item


Downloads per month over the past year

View more statistics