The thesis outlines an investigative study of the supercritical fluid extraction (SFE) of organic
natural products from two Meliaceae species. Extractions from ground wood, bark and seeds
of the selected species were studied. Qualitative and quantitative studies were performed using
a number of techniques. In the qualitative study, both off-line and on-line combination of
micro-SFE with capillary gas chromatography (SFE-GC) were performed on the wood and
bark of Cedrela toona to show the applicability of the technique in the analysis of semi-volatile
components from plant matrices. Results for the on-line and off-line SFE-GC were comparable
and indicated that the on-line studies could be used to predict results obtainable from large
scale SFE. Dynamic SFE of the limonoid (cedrelone) from the wood of Cedrela toona using
pure and modified carbon dioxide was studied on the basis of a theoretically-derived dynamic
extraction model. Extracts were collected for different lengths of time in order to investigate
the extraction kinetics of this compound. The theoretically-derived model which involved
extrapolating data obtained from shorter extraction times was used to estimate the amount of
the compound present in the wood. The estimated quantitative results were found to be
comparable to exhaustive extractions using liquid hexane as well as methanol-modified carbon
dioxide. Further, data from the SFE extractions were fitted into the model to produce a
characteristic kinetic curve from which factors controlling the extraction of the specific
compound could be predicted. Following the dynamic extraction study, a statistical
optimisation strategy was used to propose a model equation for the extraction of cedrelone
from the wood of Cedrela toona. As the preliminary step, a screening design utilising a two level,
three-factor full factorial analysis was used to study the effects of the variables pressure,
temperature and the length of extraction time on the yields of cedrelone from the wood of
Cedrela toona. By a statistical analysis of variance (ANOVA) at the 95% confidence level on
the preliminary data collected, pressure and temperature were found to be the significant
variables influencing the yields of the compound. Finally response surface methodology using
the central composite design was used in an attempt to determine the maximum response space
for the pressure and temperature effects by plotting a 3-dimensional response surface graph.
An ANOVA study at the 99% confidence level was then employed to establish the true model
equation for the extraction. / Thesis (M.Sc.)-University of Natal, 1995.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/5032 |
| Date | January 1995 |
| Creators | Modey, William Kojo. |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0025 seconds