Within plant variation of distillers dried grains with solubles (DDGS) produced from multiple raw materials in varying proportions: Chemical composition and in vitro evaluation of feeding value for ruminants

Abstract

Chemical composition and feeding value of distillers dried grains with solubles (DDGS) differ among types of raw material and ethanol production plants. Possibly more challenging, variation does as well occur between batches from the same plant. Although some comprehensive studies on within plant variation of DDGS exist, they are limited to DDGS produced from maize as the sole raw material. Nevertheless, the use of raw material blends may increase variability, particularly if proportions of raw materials vary over time. In the current study, 28 samples of blend DDGS were obtained from one ethanol plant over 420days in around 14day intervals. The raw material blend contained varying proportions of barley, maize, wheat and sugar beet syrup. Samples were analyzed in terms of chemical composition including minerals. Energy and protein values for ruminants were estimated based on chemical composition and in vitro methods. Descriptive statistics were calculated to describe within plant variation, which affected DDGS characteristics to different extents. Coefficients of variation (CV) ranged from approximately 3% for crude protein and metabolizable energy to >25% for starch, sugar and Na. Regarding protein value, predicted ruminally undegraded feed crude protein (RUP) varied considerably (CV approximately 13%), whilst in vitro results for intestinal digestibility of RUP and utilizable crude protein at the duodenum displayed consistency (CV<5%). Differing methodology impeded comparison with previous research, but results indicate that varying proportions of raw materials did not result in increased variability compared with DDGS produced from a single raw material. In addition, regression analysis was performed to examine correlations between raw material proportions and DDGS characteristics. Resulting equations have to be considered specific for the current set of samples and ethanol plant. However, equations generally reflected the proportions and ascribed quality of the raw materials. Particularly, reasonable correlations were found for proximate constituents, energy concentrations and minerals K and Mn. In contrast, other macro and trace elements either could not be predicted by raw material proportions or equations were difficult to interpret. Similarly, characteristics of the protein value displayed very limited correlation to raw material proportions. These findings were likely because minerals as well as composition, degradability and digestibility of crude protein are more prone to be affected by processing details. Thus, results suggest a limited potential to predict chemical composition of blend DDGS from raw material proportions and, on the other hand, indicate the considerable impact of processing on DDGS characteristics.