Browsing by Author "Smit, Mynhardt"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Experimental investigation of the viability of using nano-bubbles for intravascular oxygenation
    (Stellenbosch : Stellenbosch University, 2020-03) Smit, Mynhardt; Fourie, Pieter Rousseau; Pott, Robert William M.; De Jongh, Cornel; Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering.
    ENGLISH ABSTRACT: In the human body sufficient oxygen supply is of utmost importance. In certain critical conditions the lungs cannot sufficiently supply the human body subsequently other forms of oxygenation are required to potentially save the patients life. Equipment currently available for the purpose of oxygenation is often unaffordable and unavailable. The use of nano-bubbles for intra-vascular oxygenation is envisioned to possibly solve this problem. To determine if this is a viable option to be further investigated, the validation process is broken into multiple phases starting with a basic system that simulates some properties of the pulmonary system, such as flow rate, pipe diameter and systolic pressure while using water as working fluid. This thesis focussed on the first phase and serves as baseline research into the important factors associated with generating nano-bubbles using membranes. The focus of this project is to determine the effects of membrane nominal pore size, oxygen inlet pressure and acoustic interference on oxygenation of a fluid using membranes. To test the effect of these factors an experimental setup was designed, using water as working fluid, that is used to introduce oxygen bubbles into water through a membrane. Flat sheet and tubular membranes were used to introduce oxygen into the system. Two different inlet manifolds were designed to mount these membranes. Box-Behnken design was used to determine the different combinations of pore size, pressure and frequency that is to be tested and a dissolved oxygen probe was used to measure the levels of dissolved oxygen in the water associated with each combination of variables. The results obtained during experimentation was analysed, and functions were fitted to the data. The fit of the function was analysed using analysis of variance. The effect of each variable was investigated and discussed. The obtained function was then optimised using a genetic algorithm to determine the expected optimum dissolved oxygen levels. The results showed evidence that pore size, pressure and frequency influence the level of dissolved oxygen in the mixture. Using the flat sheet membrane, it was determined that the near optimum point is expected to be at 10 nm, a pressure of 2.8 bar and a frequency of 200 Hz. The effect of bubble size and zeta potential on the level of dissolved oxygen in water was investigated. It was expected that a decrease in bubble diameter would lead to an increase in dissolved oxygen levels. At a zeta potential between −9 mV and −22 mV it was observed that a decrease in bubble diameter does indeed lead to an increase in dissolved oxygen. Two sets of tests were done using the tubular ceramic membranes. Data from the sets of experiments, using tubular ceramic membranes, were combined, disregarding data obtained using the 100 nm membrane due to leakages affecting the data. It was determined that increasing the pressure, pore size and frequency lead to an increase in DO2. As the overarching aim of the project is to be able to sufficiently oxygenate a human intravenously. Considering the evidence obtained, it cannot be concluded that using membranes with nano-scale pores are a viable method for blood oxygenation, as the measured DO2 obtained in water is relatively low. The main concern, however, is the formation of large oxygen bubbles. One of the properties of nano-bubbles is that they are neutrally buoyant and that due to their negative surface charge they can stay in a mixture for long periods of time. It is hypothesised that during experimentation, nano-bubbles were generated, but remain stable in the water. As these bubbles possibly did not dissolve the dissolved oxygen levels possibly does not reflect the available oxygen. Further investigation is suggested.