F. I. Collins

Analysis of oil content of maize by wide-line NMR.

A series of experiments to define feasibility, accuracy, and precision of wide-line nuclear magnetic resonance spectroscopy as a nondestructive analytical tool for the oil content of living seeds is described. Corn samples, ranging from single seeds to 25 g, were scanned by NMR and gravimetrically analyzed. A high positive correlation (r = 0.99+) was invariably encountered. Single, 30-sec NMR scans on 25-g corn samples gave estimates within ±33 ing of the “true” oil content 95% of the time. Error associated with single, 30-sec scans of individual corn kernels amounted to ±1.3 mg of oil at the same confidence level. Samples containing more than 4.5% moisture contribute to the NMR signal, and therefore oil content is overestimated.A new dimension is added to the breeding and genetics of oil crops by the application of the NMR technique in that the process is non-destructive and feasible even for single seeds. The usefulness of the technique in studies involving the nature of water and its surrounding medium is suggested.

Analysis of oil content of soybeans by wide-line NMR

Samples of soybeans, ranging from single seed to 25 g, were scanned by NMR and then gravimetrically analyzed for oil content. High positive correlations of NMR and oil for single seed (r=0.998) and for 25-g samples (r=0.999) were found. Single 30-sec NMR scans gave accurate estimates of the oil content of sobean seeds which had been dried to less than 4% moisture content. Fifteen samples of known percentage composition, created by mixing calculated weights of soybean oil with oven dry soybean meal (made lipid-free by petroleum ether and by carbon tetrachloride extraction), were scanned by NMR. These 15 samples, starting at 2% oil and increasing by increments of 2% oil up to 30%, had a linear relationship of NMR readouts with these known percentages of oil. These results indicate that wide-line nuclear magnetic resonance spectroscopy is an accurate, rapid and nondestructive tool for determining the oil content of soybean seeds. Since NMR scanning of seed does not alter its composition or destroy its viability, this method of oil analysis could accelerate the development of new soybean strains.

Flavor evaluation of natural soybean oils of high and low linolenate content

Three varieties of soybeans, Crest, Grant, and Hawkeye, were processed in the laboratory to obtain edible oils containing 10.4, 9.4, and 5.2% linolenate, respectively. Taste panel evaluations were significantly in favor of low-linolenate soybean oils. Both high- and low-linolenate oils gave the typical off-flavors of aged soybean oil. Flavor results indicate that the linolenate content of soybean oil will probably have to be reduced below 5% to achieve a significant quality improvement in commercially processed oils. Soybean oils of excellent quality can be prepared by laboratory processing procedures.

Effect of weather damage on the chemical composition of soybeans

SummarySoybean samples containing a large proportion of damaged seed were separated into sound and damaged portions and these portions were then analyzed chemically.Analysis of these samples indicated that damage to the seed caused considerable variation in percentages of oil, protein, ash, and in iodine number and acid of the oil. Sugars in the damaged portions of the soybean samples were generally low when compared with the sound portions.Weather damaging of soybean seed caused a marked increase in percentage of crude protein. Oil percentages were sometimes higher and sometimes lower in the damaged portions. The iodine number of the oil was usually but slightly affected, and the acid number was higher in the damaged portions.In order that the chemical analysis of the samples may be representative of the true composition of a strain, the seed analyzed should be of good quality and contain no more than a small proportion of damaged seed.

Effect of relative humidity on the determination of oil in soybeans

ConclusionsFrom the results shown it may be concluded that the percent of extractable material which is obtained is highly dependent upon the atmospheric conditions under which the sample is analyzed. when soybean meals with moisture content from 4.35 to 16.8 percent are analyzed for oil content at 75–80 percent relative humidity, the amount of extractable material is not dependent upon the original moisture level. However, at lower relative humidities or lower moisture levels this is not true. Under conditions of relatively high humidity with meals of high moisture content, the short two-hour extraction gives results which check satisfactorily with the results obtained by the official four-hour method under like conditions. The data tend to emphasize the fact that the determination of oil in soybeans is empirical and that any analysis does not necessarily represent the total amount of lipids present in the sample. The data shows the necessity of control of moisture conditions under which seed is stored and under which it is analyzed if reproducible results are to be obtained.

Bauer mill operation and its effect on the percentage of oil found in soybeans

Summary and ConclusionsA Bauer Mill officially approved by the A.O.C.S. for grinding samples of soybeans to be used for oil determinations was used to grind duplicate series of the 10 check samples of soybeans sent out each year by the Smalley Foundation Committee. When the mill was operated at high temperature and forced to grind at its maximum rate, the 10 samples averaged 18.03% of oil. A duplicate series of 10 samples ground in a water-cooled mill at about one-half the maximum possible rate of grinding for the mill averaged 17.57% oil. Improper operation of a Bauer Mill when preparing soybean seed samples for analysis can cause abnormally high oil percentages. A simple mechanical device to reduce the rate of feed is suggested for use with the mill.A Bauer Mill is satisfactory for use in grinding samples of soybeans for oil analysis when it is properly operated.

Sampling soybeans for analysis

ConclusionUnder the conditions of this study, 30-gram samplings of mixtures of soybeans were found to differ significantly in oil and nitrogen content indicating the desirability of larger samples. The use of 120 to 240-gram aliquots from mixtures of soybeans which vary widely in chemical composition should tend to reduce differences due to sampling to a reasonable minimum. Differences among 30-gram samplings of highly uniform soybean seed of a single variety seem to be of slight significance. The limitations of present sampling methods should be recognized in any comparison or interpretation of chemical analyses of soybean seed.