Effects of variety and fertiliser nitrogen on alcohol yield, grain yield, starch and protein content, and protein composition of winter wheat

The effects of nitrogen (N) fertiliser on grain size and shape, starch and protein concentration, vitreosity, storage protein composition, and alcohol yield of two winter wheat varieties contrasting in endosperm texture were studied in a field trial in Herefordshire, UK in 2004. Averaged across varieties, the alcohol yield was 439 L/tonne for grain with a protein concentration of 11.5g/100g. The soft endosperm wheat variety Riband produced on average 7.7 L more alcohol per tonne of grain at a given protein concentration than the hard endosperm variety, Option. At the same time, N fertiliser was shown to have significant effects on alcohol production through its major influence on grain protein concentration. Averaged over both varieties, there was a reduction in alcohol yield of 5.7 L for each 10kg increase in protein content per tonne of grain. The starch concentration of Riband was 2.9g/100g higher than Option at a given grain protein concentration, supporting its higher observed alcohol yields. A low conversion of starch to alcohol in this study (6.30 L/10 kg starch) compared to the theoretical value (6.61 L/10 kg starch) indicated that there is potential for improvement of this character. The traits relating to grain size and shape were principally influenced by genotype, and were not influenced by N fertiliser. Conversely, there were only minor genotypic effects on grain protein concentration and vitreosity. An important finding was that there were no interactions between variety and N treatment for any of the variables considered, indicating that the response of the two varieties to changes in applied N was the same, resulting in consistent differences in starch concentration and alcohol yield between genotypes at different levels of grain protein. An analysis of the composition of the wheat storage proteins by size-exclusion chromatography showed that the gliadins increased on average by 0.56 g per g increase in total grain protein and were quantitatively the major protein fraction, suggesting that selection for low gliadin content may be a desirable means by which to reduce grain protein, and thereby increase alcohol yield in wheat. The relationship between alcohol yield per unit area and applied N rate was described by a quadratic function and the maximum alcohol yield per unit area was ca. 3630 L/ha. Statistical analysis suggested that the economic optimum rate of N applied for grain yield was close to the optimum N rate for maximum alcohol productivity. (c) 2007 Elsevier Ltd. All rights reserved.