Comparison of Corn Stover Cell Wall Polysaccharide Degradability by Rumen Microbes and a Celluosic Ethanol Conversion Process.

Release of fermentable cell wall sugars in the cellulosic ethanol conversion process is assumed similar to rumen degradability; however, available literature has only reported surrogate rumen degradation measures (dry matter, neutral detergent fiber, and fermentation gases). We determined 72-h in vitro rumen degradability of cell wall polysaccharide component sugars (glucose, galactose, mannose, xylose, arabinose, and uronic acids) and release of these sugars by a dilute acid/high temperature pretreatment and cellulase conversion process for 153 corn stover samples. Composition of the stover ranged widely (139-193, 297-384, and 171-225 g/kg dry matter for lignin, glucose, and xylose, respectively). Glucose release averaged 11% less (P < 0.001) for the conversion process (518 g/kg) than rumen degradation (583 g/kg) in a paired t-test whereas release of most other sugars was greater (23 to 65%, P < 0.001) for the conversion process. Mannose was not released by the conversion process but rumen degradability was high (820 g/kg). Pearson correlations between conversion and rumen results were moderate (r = 0.70 and 0.55 for glucose and xylose, respectively; P < 0.001), and Spearman rank correlations were of similar magnitude. Lignin concentration was negatively correlated with conversion and rumen results, respectively, for glucose (r = -0.42 and -0.47, P < 0.001) and xylose (r = -0.23 and -0.24, P < 0.01). Rumen degradability of xylose, but not release by the conversion process, was negatively correlated with ferulate cross linking (r = -0.27, P < 0.001). The higher efficiency of the conversion process for hemicellulose and pectin was expected because the pretreatment acid-catalyzed their hydrolysis. Limited correlations between the conversion process and rumen degradability of corn stover suggest that in vitro rumen degradation is not a good model for feedstock performance in a cellulosic ethanol process, although probably better than using lignification as a predictor.