Master of Science / Food Science Institute / Karen A. Schmidt / Radio frequency dielectric heating (RFDH) can provide rapid and uniform heating throughout the products' mass, and can be used to bake, dry, and defrost foods. Studies have shown that when RFDH induced a 5-log reduction of Salmonella spp. in nonfat dry milk (NDM), whey protein nitrogen index (WPNI) decreased, suggesting that functional properties of the NDM might be impacted. This research was conducted to determine if RFDH affected the functional properties of NDM [high-heat (HH) and low-heat (LH)]. Nonfat dry milk were treated to 75, 80 and 85°C in the RFDH unit, then were held for 125, 63 and 43 min for LH-NDM or 115, 52 and 43 min for HH-NDM, and cooled to ~23 ± 1°C. Powders were evaluated for WPNI , nitrogen solubility index, and color. Maillard browning and functional properties of NDM samples were evaluated after NDM was rehydrated to 3.5% protein with deionized water, and adjusted to pH 7.00. Glucono-delta-lactone was added in rehydrated NDM (3.5% protein; natural pH) as an acidifying agent to form milk gels, and the physical properties of the gels were determined. Two replications were conducted and data were analyzed with two-way ANOVA (RFDH and NDM) and Tukey mean differentiations (p ≤ 0.05). Results showed that LH-NDM (collapsed for RFDH treatments) had 5.7% less viscosity, 20.9% less overrun, 27.4% less foam stability, as well as 15% less water holding capacity compared with HH-NDM (collapsed for RFDH treatments). This can be explained by the natural whey protein denaturation differences in the HH-NDM and LH-NDM. Viscosity and surface tension were impacted by the RFDH treatment. NDM (HH and LH) treated to 85°C had 10% greater viscosity than the control, and the NDM treated to 75°C had less surface tension compared with samples treated to 80°C, 85°C and the control. Overall, RFDH decreased WPNI in LH-NDM, but not HH-NDM. The SDS-PAGE gel images provided supportive evidence to the WPNI results. RFDH is a processing technology that could change a few functional properties of NDM in this study, which makes it a promising method that may be further exploited for various food applications, such as emulsifiers, foaming agents, etc. However, potential negative impacts, such as color change caused by Maillard reaction, loss of WPNI in LH-NDM, cannot be neglected.

Probiotic dairy foods, especially non- and low-fat dairy products, are becoming popular in the US. A non-fat goat's milk yogurt containing probiotics (Lactobacillus acidophilus and Bifidobacterium spp.) was developed using heat-treated whey protein concentrate (HWPC) as a fat replacer and pectin as a thickening agent. Yogurts containing non-heat treated whey protein concentrate (WPC) and pectin as well as one with only pectin were also produced. A fat-free cow's milk yogurt with pectin was also used as a control yogurt. The yogurts were analyzed for chemical composition, water holding capacity (syneresis), microstructure, changes in pH and viscosity, mold, yeast and coliform counts, and probiotic survivability during storage at 4°C for 10 weeks. The results showed that the non-fat goat's milk yogurt made with 12% HWPC (12.5% WPC solution heated at 85°C for 30 min at pH 8.5) and 0.35% pectin, had a significantly higher viscosity (P<0.01) than any of the other yogurts and low syneresis than the goat’s yogurt with only pectin added (P<0.01). After 10 weeks in storage, viscosity and pH remained constant throughout all of the yogurts. Mold, yeast, and coliform counts were negative throughout the 10 week study. Bifidobacterium spp. remained stable and counts remained above 10⁶CFU g⁻ ¹ during the 10 week storage. However, the population of Lactobacillus acidophilus dropped below 10⁶CFU g⁻ ¹ after 2 weeks of storage. Microstructure analysis of the non - fat goat’s milk yogurt determined by scanning electron microscopy revealed that HWPC interacted with casein micelles to form a more comprehensive network in the yogurt gel. The results indicate that HWPC could be used as a fat replacer to improve the consistency of non - fat goat’s milk yogurt and other products alike.

Two studies were conducted to evaluate the effects of milk solids on restructured and emulsified turkey rolls. the milk solids used were nonfat dry milk (NFDM), whey protein concentrate (WPC), and calcium caseinate (CC). Turkey rolls consisted of 100% breast meat or 90:10 or 70:30 breast-to-thigh, salt (1%), water (10%), internal or cluster fat (10%), and 3% of various milk solids (WPC, NFDM, CC). The objectives of these studies were to 1) determine which ratio between light and dark meat is preferred; 2) determine which of milk solids evaluated will permit the highest level of dark meat incorporation into evaluated products; 3) determine if there is a mechanism by which milk proteins lighten poultry meat; and 4) determine which milk protein produces the best bind between meat pieces. Panelists were used in the first study to evaluate cooked meat attributes of color intensity, color uniformity, cohesiveness, tenderness, roasted turkey flavor, juiciness, and overall acceptability. The attributes were rated on a seven-point scale. Rolls made with WPC or NFDM scored significantly higher for color uniformity, cohesiveness, roasted turkey flavor, and overall acceptability than rolls made with CC. No differences were noted among treatments for juiciness or toughness with rolls of the same light-to-dark meat ratio. However, the 90:10 rolls were rated significantly more tender than the rolls made with the 70:30 ratio. Rolls containing milk solids had significantly higher yields than the controls. In the second study, rolls were made using the preferred meat ratio (90:10 breast:thigh meat). NFDM and WPC were used as binders, but not CC, since in the first study it was an ineffective binding agent. The second study showed that no whitening or lightening occurred in turkey rolls. This researcher also found that both NFDM and WPC increased bind strength between meat pieces. Controls made without added milk solids had less bind strength between the meat particles. Meat particle size also affected bind strength in finished products, with finely chopped rolls having higher bind strength than coarsely ground rolls. Moreover, the second study had unexpected results indicating that NFDM will prevent development of pink discoloration during refrigerated storage. The penetrometer used for bind measurements is described.

Experiment A explored the influence of sodium on direct acid, nonfat Mozzarella cheese. Cheeses with differing salt levels were obtained by varying dry salt applications (none, 0.5%, and 1.0% NaCl w/w) and hot brine stretching (0%, 5%, and 10% NaCl wt/v). Salt application and salt content influenced cheese moisture, meltability, expressible serum, micro- and ultra-structure, and color. Moisture was highest when cheese was salted before stretching (P = 0.03) . Melt was lowest in cheeses that were unsalted (P = 0.05). Cheeses stretched in salt brine had < 1% of the amount of expressible serum found in unsalted cheese (P < 0.0001). Unsalted cheeses had a more open structure with pockets of serum distributed throughout the protein matrix giving it an opaque, white appearance. Salted cheeses had a more homogeneous protein matrix lacking light scattering surfaces, resulting in a translucent cheese. Neither salt concentration nor method of salting affected the calcium content of the cheeses (P > 0.05). Experiment B explored the influence of calcium, moisture, and pH on cheese structure and functionality. Cheeses were manufactured using combinations of citric and acetic acids. Addition of EDTA to the whey during cooking, CaCl2 fortification, and extended drain times were used to produce eight cheeses in a 23 factorial design with target pH levels of 5.8 and 5.3, 70% and 66% moisture, and 0.6% and 0.3% calcium levels. EDTA was unsuccessful in removing calcium from pH 5.8 cheese. Adding CaCh successfully increased the calcium level of pH 5.3 cheese. Cheese with 0.3% calcium had greatest melt, decreased hardness and increased adhesiveness. Cheese with 0.6% calcium had decreased melt and adhesiveness, and increased hardness. When calcium content was held at 0.6% there was no significant difference in melting even when pH was varied from pH 5.8 to pH 5.3 . The microstructure of the 0.6% calcium cheeses had an increase in protein folds and serum pockets. Low calcium cheeses had a very homogeneous structure. Directly acidified nonfat Mozzarella cheese manufactured with 1.0% dry salt and hot water stretching produced the best cheese. This cheese contained 0.4% salt, 0.4% calcium, no expressible serum, and good meltability.

Properly dried sweet cream buttermilk, because of its high fat and lecithin contents, should make superior ice cream. Its use as a source of serum solids in ice cream. Its use as a source of serum solids in ice cream mix would open an outlet for this butter by-product and would serve the needs of ice cream manufacturers during nonfat dry milk solids shortage and it may make a higher score ice cream. The demand for milk solids shortage and it may make a higher score ice cream. The demand for milk solids has been increasing so rapidly that today the by-products of the dairy industry are being utilized as sources of human food to a much greater extent than before. In past years the greater bulk of creamery buttermilk has been utilized as animal feeds. Attempts are being made to convert more of this by-product into channels of human consumption. As the manufacture of sweet cream butter is increasing at a rapid pace, a larger supply of buttermilk product of high quality which is fit for human consumption is available in the market. Many previous investigators have used buttermilk products as a source of serum solids in ice cream mix with favorable results. The advantage claimed has been that it tends to improve the whipping ability of ice cream mixes and to impart richer flavor to the product. These beneficial qualities have been attributed to the butterfat and the phospholipids, of which lecithin is predominant. The work of Chapman and Supplee shows that buttermilk and cream contins several times as much licithin as skimmilk. The amounts of total phospholipids reported by Holm et al and Wright et al are approximately 1.77 per cent in dry buttermilk and 1.06 per cent in dry skim milk.

Nonfat dry milk (NOM) was evaluated to determine the effect of long term storage on its sensory quality, its solubility index and on the development of oxidative derivatives of sterols. Storage variables included the form of the milk, the atmosphere in the storage container, the type of package and the storage temperature. A trained taste panel evaluated the milk for the development of cooked, metallic, and oxidized/stale flavors, mouthfeel and overall acceptability. Gas chromatography was used to determine cholesterol content and the quantity of oxidative derivatives of cholesterol present in the samples. Although long term storage of NOM did not affect the solubility index of the milk, the sensory quality of the milk was affected. Over time storage temperature and type of package had the greatest effect on sensory quality. Milk stored for 30 months at 32 C developed an unacceptable oxidized/stale flavor and at 42 months of storage, milk stored in polybags at 21 C had also developed an oxidized/stale flavor. These results substantiate the importance of controlling storage conditions during long term storage of NOM. NOM stored for four years did not develop harmful quantities of oxidative derivatives of cholesterol. However, certain expected results did occur. NOM stored in polybags developed more oxidative derivatives than samples stored in cans. Samples stored at 32 C developed more oxidative derivatives of cholesterol than any other sample analyzed and 7S-hydroxycholesterol, the only known cytotoxic derivative of cholesterol identified, was present in the samples stored at 32 C. Fresh and control samples of NOM contained similar quantities of oxidative derivatives of cholesterol.

Two studies were conducted to determine the relative effect of various combinations of antioxidants on vitamin A oxidation and isomerization in nonfat dry milk (NDM). In the first study, one lot of pasteurized unfortified skim milk was divided, fortified with vitamins A and D and one of 11 antioxidant treatments, and spray dried. A control batch from the same lot was also fortified with vitamins A and D and spray dried. Samples were analyzed for total vitamin A bioactivity after zero, one, and two weeks of storage. After two weeks at 50°C, the only NDM samples that did not experience significant vitamin A loss were those treated with butylated hydroxytoluene, either alone at 0.57 ppm or at 0.29 ppm in combination with 250 ppm ascorbic acid. The control sample was significantly different from both of these treatments, and retained only 17% of its original retinol activity equivalents. Isomer composition changed over the two weeks of storage, with an increase of the 13-cis, 9,13-di-cis and the 9-cis isomers as well as a decrease in the all-trans isomer. In the second study, two lots of pasteurized, vitamin A & D fortified, condensed skim milk were divided into four batches, three of which were spiked with an antioxidant treatment: 250 ppm ascorbic acid + 1 ppm propyl gallate, 250 ppm ascorbic acid + 1 ppm butylated hydroxyanisole, or 2 ppm propyl gallate; the fourth batch was a control. Each of the eight batches was homogenized, spray dried and stored in the absence of light at 30°C for 3 months. Vitamin A and riboflavin were analyzed before spray drying, and after 0, 1, 2, 4, 6, 8, 10, and 12 weeks. Two treatments had significant higher vitamin A than the control, the 250 ppm ascorbic acid + 1 ppm butylated hydroxyanisole treatment and the 2 ppm propyl gallate treatment. Limited vitamin A degradation occurred in all samples during the study timeframe, although the overall degree of retinol isomerization began and remained high in all samples, with cis isomers accounting for approximately 23% of the total μg of retinol after 12 weeks. There were no significant differences in riboflavin content between any of the antioxidant treatments and no significant degradation in riboflavin over time. Antioxidants appear to be an effective means of reducing vitamin A oxidation and isomerization in nonfat dry milk. Butylated hydroxytoluene in combination with ascorbic acid was the most effective antioxidant blend observed. Antioxidants may be less effective at protecting against degradation of vitamin A if isomerization has already taken place.

Sweet cream liquid buttermilk and skimmed milk ingredients were heat processed and/or exposed to fluorescent light to determine changes in potential flavor compounds. Solid phase microextraction-gas chromatography/flame ionization detection was used to analyze the concentrations of the volatile compounds (2-butanone, 2-pentanone, acetaldehyde, diacetyl, hexanal, methyl sulfide) found in the two components. Pasteurized unoxidized skimmed milk had measurable levels of 2-butanone, acetaldehyde, and diacetyl. Pasteurization of skimmed milk increased concentration of 2-pentanone and methyl sulfide to measurable levels. However only 2-butanone and acetaldehyde were detectable in oxidized skimmed milk. All liquid buttermilk ingredient treatments had measurable concentrations of 2-butanone, 2-pentanone, and acetaldehyde. Pasteurization of unoxidized liquid buttermilk increased the concentration of diacetyl and hexanal to measurable levels whereas oxidized buttermilk, both pasteurized and unpasteurized, had measurable levels of hexanal and methyl sulfide. Nonfat (.3%) dairy beverages were formulated using the same components to determine if the volatiles in liquid buttermilk might enhance the flavor of nonfat milk. Triangle tests and hedonic tests were performed on the nonfat formulations to determine their overall difference and overall acceptance. There was no significant difference between formulations. / Master of Science