Variation in chemical composition of soybean hulls

Abstract

The objective of this study was to examine the variation in chemical composition of soybean hulls. Our goal was to develop regression equations characterizing the nutritive value of soybean hulls for use in swine diets. Samples (n = 39) were collected from different processing plants across the United States and analyzed for CP, GE, crude fiber (CF), ADF, NDF, fat, ash, Ca, P, and essential amino acids. One sample was excluded from these results because it contained approximately 10 times the amount of Ca (5.2% vs. a mean of 0.57%) as other samples. The results of chemical analysis of the samples were used to determine maximum, minimum, and mean values on a DM basis. Estimated DE values were calculated according to an equation described by Noblet and Perez (1993). Regression equations among the nutrients also were established. A high correlation was observed between CF and CP (R² = 0.92), ADF (R² = 0.96), NDF (R² = 0.97), and estimated DE (R² = 0.94), indicating that the analyzed fiber content of soybean hulls could be used to predict the other components. A high correlation also was observed between CP and estimated DE (R² = 0.90). Lower correlations were observed between ash concentration and Ca and P. High correlations were observed between CP and lysine (R2 = 0.89), methionine (R2= 0.88), threonine (R² = 0.93), and tryptophan (R² = 0.93). In summary, the chemical composition of soybean hulls can be highly variable; however, CF content can help explain much of the variation in CP, ADF, NDF, and estimated DE, and CP content can be used to predict individual amino acid levels.