Freshness is a holistic attribute of a food product, with a meaning that most often includes how recently produced or harvested a food currently is and to what extent it has been preserved. Without knowledge of production date, consumers will judge perceived freshness by the sensory properties of foods. However, these properties are not easily identified and described, as they vary considerably between different product types. The objective of this thesis, was to determine the sensory properties that consumers most associate with baked product freshness, and to determine the volatile composition responsible for "fresh-baked" flavour. In addition, relationships were determined and modelled between volatile composition, sensory properties and perceived freshness, in order to predict perceived freshness without the need for further consumer testing. Two consumer studies were carried, where perceived freshness of 10 breads, 5 cakes and 5 biscuits was measured based on appearance, flavour and texture, and by product odour only. Individual differences in perceived freshness were represented on two-dimensional Freshness Maps generated using Principal Component Analysis (PCA). Consumers used similar terms to describe product freshness, but different terms were associated with specific product types. An understanding of the sensory characteristics associated with perceived freshness was determined by examining relationships between perceived freshness and descriptive analysis (DA). Sensory characteristics perceived to convey freshness in one product type did not necessarily convey freshness in another. Further relationships between perceived freshness and sensory characteristics were determined for 20 bread types. Descriptive analysis was carried out with all breads, whereas perceived freshness was measured for 10 of these. Three consumer clusters were identified that were homogenous in their freshness perceptions, indicating that perceived bread freshness varied among consumers. Perceived freshness of breads not evaluated by consumers, but assessed by descriptive sensory analysis, were predicted for each consumer cluster using Partial Least Squares Regression (PLSR). Most fresh breads were described as having "porous" appearance, "floury", "toasted" and "malty" odour, "sweet", "buttery", and "oily" flavour, and "sweet" aftertaste. Least fresh breads were described as "musty" odour, "sour" flavour and "sour" aftertaste. In addition, Proton Transfer Reaction Mass Spectrometry (PTR-MS) measured the volatile composition of all 20 bread types. Thirty-three mass ions significantly discriminated between bread types. Relationships were determined between odour and flavour sensory attributes, and mass spectral data using 15 of the 20 breads. PLSR models predicted the sensory properties and perceived freshness of all breads including five types not included in the construction of models. Bread perceived to be most fresh shared a similar combination of positively correlated mass ions, m/z 87, 97 and 117, also represented by "dairy" odour and "buttery" flavour, whilst the masses m/z 63, 69 and 91 were negatively associated with bread freshness, represented by the sensory attributes "grain", and "musty" odour, and "sour" flavour. The influence of sweetener, fat type and time from baking on perceived freshness was determined for cakes. Four sweeteners (sucrose, glucose, xylitol, isomaltose), three fat types (butter, margarine, shortening) and two times from baking (days zero and 15) were investigated. Descriptive analysis was carried out using a trained panel, and volatile composition of measured using PTR-MS. Twelve cake samples with the largest sensory variation were further selected for consumer testing for perceived freshness. Results demonstrated that sweetener type, fat type, time from baking and their interaction significantly influenced the sensory properties, volatile composition, and perceived freshness of cakes. There was no evidence of perceived freshness segmentation among consumers. Perceived freshness of the twelve cakes not evaluated by consumers was predicted using sensory properties, volatile composition, and their combination. The freshest cakes were evaluated on the day of baking, and contained a combination of margarine and sucrose, and butter and sucrose. The least fresh cakes were evaluated after 15 days of storage and contained a combination of shortening and isomaltose, and butter and isomaltose. The most fresh samples were positively correlated with mass ions m/z 124, 74, 97 and 93, and sensory characteristics of "buttery" odour, and "buttery", "eggy" flavour. The least fresh samples were negatively correlated with mass ions m/z 110 and 95, and sensory characteristics of "rancid", "dusty" and "fatty" odour. Best predictions were obtained for cakes evaluated on day zero that contained a combination of shortening and sucrose. This research established an objective knowledge of consumers' perceived freshness of baked product types in terms of sensory properties and volatile composition. This approach enabled the understanding of ingredients and time from baking influences, as well their interaction on baked product freshness. The predictive models developed that examined relationships between PTR-MS spectra, sensory characteristics and consumer perceived freshness can be applied to predict freshness of baked products not assessed by consumers.

The effects of the fortification of tortilla chips with mechanically expelled soy flour as well as baking and frying processes on the properties of tortilla chips were evaluated. Sensory characteristics, texture, thickness, color, protein and oil content were evaluated. Texture was measured by objective and subjective tests. Sensory properties were evaluated using a nine point hedonic scale. Soybeans (food grade Hartz) were mechanically expelled to obtain partially defatted soy flour of 6.7% final oil content. Dry masa flour (DMF) was replaced with 0, 10, 20 and 30% mechanically-expelled soy flour (MESF). The equilibrated tortilla was either fried in oil or baked in an air-impingement oven followed by convection oven drying. Overall, fried tortilla chips were harder and thicker than baked tortilla chips. Fried tortilla chips with 20 and 30% soy flour substitution required less force to break. In fried tortilla chips, as MESF increased, force and work levels decreased, where 20% MESF had the lowest force values. Thickness measurements of tortilla chips showed that as the thickness increased, the force and work also increased. Protein increased linearly in baked and fried tortilla chips where 30% resulted in the highest protein level. In fried tortilla chips, MESF fortification increased oil levels linearly as well. Baked tortilla chips were lighter than fried tortilla chips. In a consumer sensory evaluation, fried tortilla chips were preferred more than the baked ones. In fried tortilla chips, 20% had the highest sensory scores overall. Ten and 20% MESF fortification in fried tortilla chips were the most acceptable of all. In all treatments, regardless of type of processing, panelists could not detect any “beany” flavors in any of the samples. Therefore, dry extrusion followed by mechanical expelling proved successful in creating a suitable soy flour for tortilla chip production. MESF can be added at 10-30% levels in tortilla chips. Up to 20% would be recommended. Frying results in higher acceptability consumer scores over baking.

In this research, initial studies were done to determine the effect of various methods of presentation involving chemical preservatives, water activity ($ rm a sb{w}$), and modified atmosphere packaging (MAP) on mold growth in an agar model system. Results showed that preservatives could completely inhibit mold growth for 2-40d depending on concentration and pH used. Gas packaging (60% or 80% CO$ sb2$), oxygen absorbents, alone or in combination with potassium sorbate, could also inhibit mold growth for $>$40d at ambient storage temperature using a Response Surface Methodology (RSM) approach. / The effects of various methods of applying potassium sorbate into pizza crusts via direct incorporation into the batter, surface spraying, and impregnation of packaging material with potassium sorbate to control mold spoilage of pizza crusts were also investigated. Results showed that the antimicrobial effect of potassium sorbate was negligible when the packaging material was impregnated with the inhibitor but more pronounced when it was incorporated directly into the dough or sprayed onto the product's surface. The inhibitory effect of potassium sorbate increased as both the pH and the inoculum level decreased. / Shelf life studies using low concentrations of potassium sorbate (1000 and 2000 p.p.m.) and MAP, alone and in combination with each other, showed that potassium sorbate, gas packaging or oxygen absorbents (Ageless FX) could extend the shelf life of pizza crusts and decrease the growth rate of molds, bacteria and yeast. Furthermore, when pizza crusts were packaged in 60% CO$ sb2$ or with an oxygen absorbent, in combination with potassium sorbate (1000-2000 p.p.m.), a shelf life of 42d was possible without compromising the sensory shelf life of the product. (Abstract shortened by UMI.)

Thesis (Ph. D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2005. / Dr. Jennifer E. Michaels, Committee Chair ; Dr. Bonnie Heck Ferri, Committee Member ; Dr. George J. Vachtsevanos, Committee Member ; Dr. Magnus Egerstedt, Committee Member ; Dr. Farrokh, Ayazi, Committee Member ; Dr. Sheldon M. Jeter, Committee Member. Vita. Includes bibliographical references.

Thesis (M.S.)--University of Missouri-Columbia, 2004. / Typescript. Includes bibliographical references (leaves 203-218). Also available on the Internet.

The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Title from PDF of title page (University of Missouri--Columbia, viewed on September 24, 2009). Thesis advisor: Dr. Fu-Hung Hsieh. Includes bibliographical references.

Thesis (M.S.)--University of Missouri-Columbia, 2004. / Typescript. Includes bibliographical references (leaves 203-218). Also available on the Internet.

No description available.

No description available.

M.S.