University of Kansas
Phase Equilibria and Transport Properties of Gas-Saturated Ionic Liquids
Abstract
dc:description.abstractIonic liquids have emerged as promising solvents for gas capture and separation technologies over the past two decades. Composed entirely of ions with melting points below 100 °C, these liquids offer significant potential to replace conventional solvents as more efficient and environmentally friendly alternatives. A vast range of ionic liquids can be created through different combinations of cations and anions or by mixing ionic liquids, offering high tunability of their thermophysical properties to meet specific needs. However, assessing their viability as alternative solvents in various applications requires a comprehensive understanding of these thermophysical properties both as pure components and in solution. Growing awareness of the environmental impacts of fluorinated gases (F-gases) has led to stringent regulations mandating the phase-out of high global warming potential (GWP) F-gases. Hydrofluorocarbons (HFCs), a particularly potent class of F-gases widely used in refrigeration and other industries, are often found in azeotropic mixtures that make their separation challenging. Extractive distillation with ionic liquids (ILs) as entrainers has emerged as a promising alternative for effective HFC separation. To evaluate the feasibility and optimize the design of such processes, a comprehensive understanding of the thermophysical properties of relevant ILs, HFCs, and their mixtures is essential. Although solubilities and selectivities of HFC/IL mixtures have been extensively studied, data on other thermodynamic and transport properties remain limited. In this study, we investigate key thermodynamic and transport properties of HFC/IL mixtures, including solubility, density, viscosity, interfacial tension, thermal conductivity, electrical conductivity, and self-diffusivity. Accurate knowledge of component and mixture densities is crucial for several design parameters, including heat and mass transfer coefficients. In this study, we employ a custom-built apparatus capable of simultaneous vapor-liquid equilibrium solubility and density measurements to determine mixture compositions and densities. Our findings show that the molar volumes of all examined HFC/IL mixtures decrease with increasing solute concentration, with large, negative excess molar volumes observed across all mixtures. Supported by spectroscopic data from this work and previous MD simulations, we demonstrate that dissolution is primarily influenced by free-volume effects. However, we reveal that both the distribution and total free volume are key factors, and that free-volume distributions in flexible ILs exhibit dynamic behavior. The inherently slow dynamics of ILs often limit their effectiveness as alternative solvents in various processes. Additionally, a theoretical understanding of IL-based solution dynamics remains limited. Here, we demonstrate that introducing small amounts of HFCs significantly reduces the viscosity of several imidazolium-based ILs and progressively enhances self-diffusivities at low to moderate compositions. However, these systems frequently show considerable deviations from simple Stokes-Einstein hydrodynamic behavior. Due to the complex nature of IL-based systems, conventional theoretical models for solvent dynamics inadequately capture the unique behavior of these ILs and IL-based mixtures. Through a combination of spectroscopic and scattering techniques, supported by molecular dynamics simulations, we propose that configurational and conformational changes along ionic charge networks as well as electrostatics play a critical role in the anomalous dynamics observed, despite the persistence of charge ordering. We show that viscosity mixture models utilized in common process simulations software such as Aspen Plus result in large deviations that translate to significant errors in unit design. Extensions of semi-theoretical models such as the Eyring-NRTL and Allal FVT models are proposed here for improved property predictions. Thermal conductivity investigations of HFC/IL mixtures conducted here reveal significant deviations from simple mixing behavior. The thermal conductivity remains largely governed by the pure IL up to high HFC concentrations, beyond which an abrupt decrease can occur. Spectroscopic and scattering analyses suggest that, unlike overall dynamics, thermal energy conduction is predominantly controlled by charge network percolation. Existing theoretical and empirical models fail to accurately capture this behavior, leading to substantial implications for unit design, including heat exchanger sizing. Interfacial tension values of HFC/IL mixtures are critical for estimating mass transfer coefficients in extractive distillation columns and heat transfer coefficients in heat exchangers for any HFC/IL application. In this study, we design and employ a custom pendant drop apparatus to measure interfacial tension in various HFC/IL mixtures, revealing that values can exhibit positive or negative deviations from simple mixing depending on the HFC and IL combination. We propose that the IL molecules preferentially orient their apolar groups toward the interface and polar domains toward the bulk. The size, symmetry, and polarity of the HFCs influence the extent of their interfacial adsorption and the IL orientation, ultimately affecting the overall interfacial tension. The molecular-level dynamics and structuring in ILs remain poorly understood, and theoretical descriptions of IL-based mixtures are even less developed. In this work, we combine spectroscopic and scattering techniques with molecular dynamics simulations conducted by our collaborators to elucidate the molecular structuring and interactions responsible for the anomalous thermophysical behavior of complex HFC/IL mixtures. Our findings reveal that these complex interactions often lead to substantial deviations from conventional solvent modeling techniques, resulting in significant errors in unit design, such as mispredictions in heat and mass transfer coefficients. Such inaccuracies can lead to oversized or undersized equipment, increasing capital and operational costs and, in some cases, causing unit underperformance. To enhance property modeling and process simulations, we propose empirical extensions of semi-theoretical models for properties like density and viscosity.
Degree
thesis:*- Grantor dc:publisher
- University of Kansas
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Al-Barghouti, Karim Samir
- Advisor dc:contributor.advisor
-
- Scurto, Aaron M.
Subjects
dc:subject × 7Rights
dc:rights- Statement dc:rights
-
- This item is protected by copyright and unless otherwise specified the copyright of this thesis/dissertation is held by the author.
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Other
- https://www.proquest.com/LegacyDocView/DISSNUM/31766473
- OAI identifier oai:identifier
- oai:kuscholarworks.ku.edu:1808/38022