Using computational models to understand CO2-rich fluid behaviour, photochemical reactions and high-temperature water reactions.

Wiseall, Christopher (2020) Using computational models to understand CO2-rich fluid behaviour, photochemical reactions and high-temperature water reactions. EngD thesis, University of Nottingham.

[img] PDF (Chris Wiseall EngD Thesis) (Thesis - as examined) - Repository staff only until 24 July 2022. Subsequently available to Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (7MB)

Abstract

This Thesis uses computational model development throughout to aid with the understanding of experimental results in two general areas: understanding phase behaviour of CO2-rich fluids and modelling of novel chemical reactions.

Chapter 1 the introduces experimental measurements that were carried out to measure the solubility of water in a 10% H2 + CO2 fluid mixture at 40 °C and above 8 MPa. The experimental results showed a continuation of the trend observed previously with pure CO2, 5% N2 + CO2 and 10% N2 + CO2, where the solubility of water was dependent on the density of the mixture. The experimental results carried out in this work, along with the previously reported results, including at 25 °C, were compared to those predicted by both the Advanced Peng-Robinson 1978 and SAFT-γ Mie Equations of State. The predicted water solubility was generally underestimated compared to experimental results, which was considered to be safer than an overestimation for any future pipeline network.

The minimal saturation limit was ca. 1500 ppm, which with a safety multiplier of 60%, suggested that water solubility limits of 1000 ppm lay above the previously suggested limits of 500 ppm. To be surer of this, experiments would need to be continued at lower temperatures, such as 10 °C, which are more likely to be found as pipeline operating conditions, especially in the UK.

A computational model of a pipeline was then used to investigate the effect of impurities on pipeline behaviours, such as pressure drop, which is a crucial measure in the development of a pipeline network. The presence of impurities was shown to have an effect on the pressure drop of a pipeline, with the identity of the impurity exhibiting a much less pronounced effect.

The pipeline model was then adapted to a smaller size to support a technique previously developed within the group to measure the phase behaviour of reactive mixtures, using an increased in pressure drop to indicate a phase change.

Within Chapter 2 both photochemistry and high-temperature water reactions are introduced before the process modelling and the mathematical techniques used throughout this Thesis are introduced in Chapter 3.

The models developed for photochemical reactions and photochemical reactors are introduced in Chapter 4. Although previously reported models of photochemical reactions were limited, the model developed in this work was shown to agree well. Maximum likelihood estimation was then used with the model to estimate the quantum yield of a number of reactions, assuming the lamp manufacturer’s light intensity was accurate. The quantum yields were then used to accurately predict further experimental results, outside of the experimental space used in the estimation.

The photochemical reactor models were then developed for the FEP tubular reactor and Vortex reactors, which were much more complex than the excimer flow and batch reactors. The Vortex reactor process model matched the Computational Fluid Dynamics model well, with the ability to predict the performance of reactions in the Vortex reactor.

Within Chapter 5 a high-temperature water reactor model was developed and used to model a multi-step reaction in two different reactors; a large-scale and small-scale reactor. Maximum likelihood estimation was used to estimate the kinetic parameters of the reactions, with a reparameterization of the Arrhenius equation and an optimisation algorithm used to reduce the correlation and uncertainty of parameters. The estimated kinetic parameters from the smaller reactor were then used to predict the performance of the larger reactor, which was a key aim of the model development.

The thermal reaction modelled in Chapter 5 was the second stage of a reaction modelled in the photochemical chapter. The daisy-chaining of these two reactions was then modelled with a possible equivalence of 2.5 kg per day being possible if the laboratory equipment was available. A future reactor was then designed using the reaction’s kinetics to optimise the yield, while reducing energy and material use.

Item Type: Thesis (University of Nottingham only) (EngD)
Supervisors: George, Michael
Poliakoff, Martyn
Keywords: Carbon capture and storage, Photochemistry, High-temperature water, Chemical engineering, Process engineering, Chemistry, Quantum yield, Parameter estimation, Pipelines, Reactor design
Subjects: Q Science > QD Chemistry > QD450 Physical and theoretical chemistry
Faculties/Schools: UK Campuses > Faculty of Engineering > Department of Chemical and Environmental Engineering
UK Campuses > Faculty of Science > School of Chemistry
Item ID: 60501
Depositing User: Wiseall, Christopher
Date Deposited: 31 Jul 2020 04:40
Last Modified: 14 Aug 2020 09:07
URI: http://eprints.nottingham.ac.uk/id/eprint/60501

Actions (Archive Staff Only)

Edit View Edit View