Browsing by Author "Wong, Marie"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- ItemCellular changes during cold-pressed ‘Hass’ avocado oil extraction : a thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy in Food Technology, at Massey University, Auckland campus, New Zealand(Massey University, 2019) Yang, ShuoCellular changes during cold-pressed extraction of ‘Hass’ avocado (Persia americana Mill.) and ‘J5’ olive (Olea europaea L.) were investigated to understand how each step in the process affects oil release from the tissue and to ascertain if and how cold-pressed oil yields were influenced by cellular changes. Electrical impedance spectroscopy (EIS), electrical conductivity, light microscopy and rheological measurements were used to examine avocado and olive flesh and pulp structure at defined processing steps during both commercial and laboratory-based cold-pressed oil extraction. Light microscopy revealed most parenchyma cells in the fruit flesh were ruptured after the destoning and grinding steps. Concomitantly, a significant reduction in electrical resistance and a concurrent increase in conductivity of pulp tissue occurred when cells were ruptured during the destoning and grinding process. Malaxing assisted aggregation of oil into larger droplets, observed by microscopy. Increasing malaxing time resulted in a decrease in the solid-like behaviour (G’) of fruit pulp and an increase in cold-pressed oil yield, which correlated with the oil droplets in the fruit paste coalescing together and becoming larger. Idioblast cells in avocado flesh appeared to remain unruptured and intact during the extraction process. In comparison to the cold-pressed oil extraction of ‘Hass’ avocado, olive oil was easier to recover from ‘J5’ olive during cold-pressed extraction (at lower temperatures and for shorter times) as the olive paste was less viscous allowing the oil droplets to aggregate more easily. Processing of avocado fruit at three different stages of ripening (minimally-, fully- and over-ripe) produced higher oil yields and decreased oil quality (based on % free fatty acids and peroxide value) with riper fruit. Intact fruit and fruit pulp from the over-ripe fruit had higher conductivity and lower electrical resistance values, which indicated more cell rupture occurred when softer, riper avocado fruit are processed. For avocado fruit at six different stages of maturity (harvested between September and April during the 2016/17 season), light microscopy results showed there were more unbroken parenchyma cells in early season, less mature fruit. Polysaccharides in the cell walls were more strongly bound to cellulose in early-season avocado fruit. Late season fruit had more cell disruption during extraction corresponding to higher conductivity and lower electrical resistance values; hence higher extraction yields with increasing maturity. No significant compositional changes of the polysaccharides in the cell walls occurred during malaxing, which indicated that the malaxing step only promoted aggregation of the oil droplets. The malaxing temperature and ultrasound treatment at 20–25 kHz did not assist with cellular disruption during extraction. Higher malaxing temperature reduced the viscosity and increased the mobility of oil droplets and oil droplets were more likely to collide and aggregate to form larger droplets, reducing the G’ of the pulp tissue. The oil yield significantly increased from 1.05% to 13.43% with malaxing temperature increasing from 30 to 50 ⁰C, for early season fruit. Ultrasound treatment at 20–25 kHz decreased the G’ of the avocado pulp and helped the oil to aggregate. In conclusion, the avocado flesh cellular structure ruptured more easily in softer and late maturity fruit contributing to increased oil yields. Malaxing time, temperature and ultrasound treatment at 20–25 kHz influenced the degree of oil aggregation in fruit pulp and therefore improved the cold-pressed oil yield. Olive pulp was less viscous or less solid like during malaxing, resulted in faster oil agglomeration.
- ItemModelling of a direct osmotic concentration membrane system : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Food Engineering at Massey University(Massey University, 1997) Wong, MarieDirect osmotic concentration (DOC) is a novel continuous membrane process. Two co-current streams, separated by a semi-permeable membrane, are recycled through a DOC module. The turbulent-flow dilute juice stream is concentrated by osmotically extracting water across the membrane into a laminar-flow, concentrated osmotic agent (OA) stream. The semi-permeable membrane is asymmetric, with a non-porous active layer (15 μm) and a porous support layer (150 μm). Membrane solute rejection was greater than 99%. Normal operation orients the active layer towards the juice stream. For this study, water (osmotic pressure = 0) was used in the juice channel. The relationship between water flux rate and the osmotic pressure of the bulk OA stream was asymptotic, reaching a maximum flux of 1.75 x 10-3 kg m-2 s-1, when using fructose OA at 15 MPa osmotic pressure and 20°C. Flux rates doubled when NaCl replaced fructose as OA. A doubling in temperature to 40°C resulted in a 50% increase in flux rate. OA solution properties, particularly viscosity and factors affecting diffusion coefficients had a strong influence on flux rates. When the membrane was reversed, with the active layer facing the OA channel and the support layer filled only with water, flux rates were 40 to 60% higher than the normal orientation. There were three resistances to water flow associated with: osmosis across the membrane active layer (R1), diffusion and porous flow across the support layer (R2), and; diffusion across the boundary layer in the OA channel (R3). For fructose OA at 0.50 g (g solution)-1 (osmotic pressure = 15 MPa), R1 contributed 9% of the total resistance to water flux in the DOC module, R2 contributed 64% and R3 contributed 27%. For an iso-osmotic concentration of NaCl OA (0.15 g (g solution)-1) the relative resistances were. R1 = 17%, R2 = 44% and R3 = 39%. It was clear that the water flux from the dilute to concentrated stream was more strongly influenced by the support membrane and OA solution properties than the active semi-permeable membrane itself. This accounted for the asymptotic relationship between bulk OA stream properties and flux rate. The mathematical model successfully incorporated these resistances and solution properties. Data calculated using this model agreed well with experimental results.