A. J. Clifford

Occurrence of cyclic fatty acid monomers in frying oils used for fast foods

Cyclic fatty acid monomers were analyzed by gas chromatography in commercial frying oils, obtained in this country and in the Middle East. Samples were obtained from food outlets in California and Illinois after varying periods of usage. The samples from Egypt and Israel were collected from street vendors frying vegetable patties (known as “fallafel”) in open-air stands. The United States samples ranged from 0.1 to 0.5% cyclic monomers, and from 1 to 8% polar +noneluted thermal oxidation materials. The Middle Eastern samples showed significantly more heat abuse, with values for cyclic monomers from 0.2 to 0.7% and polar materials ranging from 2 to 22%.

Changes in physical and chemical properties of shortenings used for commercial deep-fat frying

This study evaluated some of the changes that occur in shortenings used for commercial deep-fat frying in fast-service restaurants. Foods cooked in partially hydrogenated soybean oil were battered chicken parts and french fries. Sixty-five samples of fresh and used shortenings were collected from nine restaurants on three occasions over a three-month period. Frying periods varied from 0 to 300 hr, and most samples were taken just before the used fat was discarded. For fresh shortenings, percentages of polar materials, free fatty acids (FFA), materials not eluted by gas chromatography, and fatty acid profiles differed only slightly. For used samples, there were marked variations in these analyses and in increases of dielectric constant measurements. Frying times were highly correlated with increases in dielectric constant, polar materials and FFA. The greatest change in fatty acid profiles occurred intrans-C18 monoenes which decreased from over 40% to as low as 13%. Due to lipid exchange with chicken fat, both oleic and linoleic acids increased in the shortenings with hours of use, whereas stearic acid decreased. There were high correlations, among increases in dielectric constant, percentages of polar materials and FFA, demonstrating that each of these methods could predict degradation of the shortening. However, the increase in dielectric constant, as measured by a Foodoil Sensor (FOS), was the most convenient for quality control in restaurant situations. In most cases, used shortening was discarded before 100 hr of frying time; and only a few of these samples had FOS readings near 4.0, FFA over 1.00%, or percentages of polar materials over 27%. These values have been suggested as discard criteria. However, a number of samples used between 100 and 300 hr exceeded these limits. There is a need to specify suitable limits, related to quality and health factors, to determine at what point a cooking fat should be discarded.

Lipid content and fatty acid profiles of various deep-fat fried foods

Forty-one brands of nine different types of snack and convenience foods were purchased from food stores and fast service restaurants in the Sacramento area of California. All samples had been prepared by deep-fat frying. They included potato chips, corn chips, tortilla chips, cheese chips, cheese puffs, cake donuts, french fries, chicken pieces and fish pieces. These samples were analyzed in duplicate for total fat and fatty acid composition. The total lipid content of each type of food varied among different commercial sources; the average percentages were as follows: potato chips, 40; cheese puffs, 38; corn chips, 35; cheese chips, 25; tortilla chips, 24; cake donuts, 22; chicken thighs, 14; french fried potatoes, 14, and fish pieces, 10. The fatty acid profiles of the total lipids in several brands of potato chips were relatively constant. The fatty acid profiles of the total lipids in the corn and cheese snack foods varied widely. Fatty acid compositions of donuts, chicken and fish pieces and french fries were influenced by the amount and fatty acid profile of the lipids in each uncooked food, as well as by the composition of the cooking fat.

Estimating Derivatives of Pharmacokinetic Response Curves with Varying Bandwidths

A kernel-smoothing method with locally varying bandwidths for the nonparametric estimation of derivatives of a function is proposed for highly nonequidistant data as they occur in pharmacokinetic response curves. We construct estimates having the particular property that the derivative estimates correspond exactly to the corresponding derivatives of the curve estimate even under locally varying bandwidth choice. The effects on the estimation of peak location (characteristic points) are investigated. In an example, characteristic points are estimated for a recently developed in vivo isotope dilution assay for vitamin A (retinol) nutritional status. The in vivo kinetics of appearance and disappearance of isotopically labeled retinol can be described with the proposed method.